The fight-or-flight response (FFR) is an instantaneous organismic response driven by emotions to external stimuli. However, most current intelligent systems requiring real-time environmental response overlooked this fundamental mechanism. Meanwhile, endowing systems with pre-response adaptive decision-making ability based on environment complexity (EC) considerations also necessitates in-depth research. This work proposes a decision-making model with a bionic memristive circuit comprising n FFR circuits and one environment enrichment (EE) module. The FFR circuit simulates FFR and implements long-term emotion memory (LTEM), memory forgetting, emotion generalization and fast emotion arousal. The EE module allows the circuit to dynamically adjust response targets in multi-stimulus scenarios by considering EC. Consequently, the circuit achieves adaptive and minimal delay responses to changeable stimuli via pre-response decision-making, validated by PSPICE simulations. The use of memristors facilitates online in-situ operating and in-memory computing, which makes the circuit expected to be deployed on multi-nozzle fire-fighting robots, enabling them to adaptively prioritize target processing in real time based on the urgency of various targets in multi-fire points scenarios.

Medical institutions store a vast amount of patient information, and various types of medical data face severe security challenges in cloud storage environments. This paper proposes an efficient multi-type medical multimedia data encryption scheme based on the memristive Hopfield neural network (MHNN). First, a class of MHNNs is constructed, which exhibits isomorphic extreme multistability under different initial conditions. The systems are capable of generating large-scale coexisting chaotic attractors, whose spatial positions, vortex numbers, and amplitudes can be independently regulated, thereby significantly expanding the generation capacity and diversity of chaotic sequences. Based on this, a multi-type data encryption algorithm is designed, which uniformly encodes and integrates different types of medical data into a secure transmission structure. Furthermore, by exploiting multistable characteristics of the MHNN, a two-layer key system composed of a master key and a selection key is established, enabling a secure medical data scheme that supports multi-party collaborative annotation. Experimental results demonstrate that the proposed encryption scheme achieves excellent performance in both security and efficiency, providing a novel and effective solution for secure storage and collaborative processing of multi-type medical multimedia data.

Existing computational methods for the $k$ -winners-take-all ( $k$ WTA) operations often suffer from limitations in eliminating lagging errors, high computational complexity, and weak robustness. To deal with these challenges, we propose an integral-enhanced adaptive gradient neural network (IAGNN) for $k$ WTA. We demonstrate that the IAGNN integrates an adaptive coefficient to eliminate lagging errors while retaining an $O(n^{2})$ computational complexity. We proved the Lyapunov stability and robustness of the IAGNN. We provide a numerical simulation, and the results demonstrate the stability and robustness of the IAGNN. Furthermore, we implement the IAGNN in a multirobot tracking system for competitive allocation coordination, and the results demonstrate the operational feasibility and noise resistance of the IAGNN.

For the frequency regulation and power allocation problem for ac microgrids, the multiple constraints of load frequency restrictions, nodal active power injections, and power-flow balance that guarantee transient stability are as crucial as the final consistent steady-state results. This article investigates a class of droop-controlled ac microgrids with the abovementioned constraints, and it presents sufficient and necessary conditions to improve system robustness and reliability under load fluctuations. By integrating the Kuramoto oscillator model into the primary control law, cyber-physical coupling dynamics for ac microgrids are established. A novel communication-network-based secondary distributed control approach is presented, which considers physical nodes with different characteristics of power generation and load units. By using invariant theory and nonquadratic Lyapunov techniques, the bounded input-output stability of microgrids can be guaranteed under certain conditions, which is especially pertinent in active power allocation. The theoretical results are verified through numerical case studies of both public and self-built power test systems, demonstrating the obvious improvement in robustness and reliability against variable power generation and load demand.

The growing integration of massive multi-source heterogeneous distributed generation (DG) into power grids and the emergence of cluster-collaboration-based generation modes have highlighted significant challenges in effectively controlling microgrid clusters (MGCs). This paper develops an interval type-2 (IT2) Takagi-Sugeno (T-S) fuzzy model for the secondary control layer of MGCs with additive disturbances, designed to capture the dynamics of heterogeneous DG under non-cyber attacks. By adopting the constrained-tightening-based prescribed performance control strategy, the proposed low-computation distributed fuzzy secondary control protocol achieves voltage cooperative restoration for MGCs subject to additive disturbances. Notably, this protocol eliminates reliance on the dynamics of upper-level DG for lower-level communication networks while the efficient multi-layer communication structure is employed to facilitate cluster collaboration. Subsequently, the MGC security detection-isolation-deferred switching framework under distributed cyber attacks is established. The proposed method based on performance monitoring for security detection, node isolation, and performance recovery, demonstrating that multi-source heterogeneous DGs can be reconnected to the MGC through deferred switching. This approach enhances the feasibility of the design while maintaining MGC performance standards. Finally, a standard IEEE 33-node distribution model is employed to validate the proposed control protocol's effectiveness via the OPAL-RT real-time simulation platform.

The existing group formation-containment (GFC) studies for multi-agent systems (MASs) depend on system model information and generally neglect input saturation constraint, thereby limiting their applicability to MASs with unknown system model and input saturation. This paper investigates the data-driven GFC control problem of nonlinear MASs with asymmetric input saturation. First, a novel communication topology selection algorithm with relaxed topology conditions is proposed. Then, to tackle the challenge posed by asymmetric input saturation, a novel nonquadratic performance index function with a simplified formulation is designed and the corresponding Hamilton-Jacobi-Bellman equation is derived. On this basis, an effective value iteration algorithm is proposed to determine the optimal GFC control policy, accompanied by the rigorous mathematical analysis. By establishing the critic-actor framework based on generalized fuzzy hyperbolic model, a novel data-driven algorithm is proposed to achieve GFC under asymmetric input saturation, which overcomes the dependence on system model. Finally, some simulation results are provided to verify the effectiveness and superiority of the proposed data-driven GFC algorithm.

This paper develops a novel dynamic sampled-data-based event-triggered (DSET) secure fuzzy tracking control scheme for a category of unknown nonlinear heterogeneous multiagent systems (MASs). Asynchronous DSET mechanisms are developed, in which the triggering thresholds evolve dynamically with the system states. The proposed mechanisms only require discrete state sampling and triggering evaluation, thereby avoiding continuous monitoring and reducing triggering frequency. Moreover, the triggering interval is shown to admit a positive lower bound that is not solely determined by the sampling period. Under denial-of-service (DoS) attacks, distributed DSET observers are constructed to estimate the leader's system matrices and states, where only a subset of followers requires access to the leader information. A detection scheme is further designed to identify the termination of DoS attacks. Furthermore, a fuzzy fault-tolerant control scheme is proposed based solely on output measurements. The convergence of observations and tracking errors is rigorously established. Simulation results demonstrate the effectiveness and superiority of the proposed approach.

The widespread deployment of smart meters has created significant opportunities for applying artificial intelligence technologies to power system tasks. However, the high cost of data annotation limits the effectiveness of traditional supervised learning in this domain, making self-supervised learning an attractive alternative. In this article, we propose PowerDiffuser, a novel self-supervised learning strategy tailored for power load signals. By leveraging a diffusion model framework, PowerDiffuser integrates two mainstream self-supervised paradigms, namely contrastive learning and reconstruction-based learning, which enables the model to effectively capture both periodic patterns and local features. To address the overfitting issues commonly observed in generic time-series feature extractors when applied to power load tasks, we design two modular spatiotemporal feature extractors specifically engineered to handle samples with varying complexity levels. In addition, we adapt the involution operator to better align with the unique characteristics of power load signals. Extensive experiments on the ISMCBT, ETTh and REDD datasets demonstrate that PowerDiffuser consistently outperforms both time-series general models and existing self-supervised learning strategies across diverse downstream power load tasks. Ablation studies further validate the contributions of the proposed modules and highlight the effectiveness of transforming 1-D load signals into 2-D periodicity-based representations as a preprocessing step.

This article proposes a multiagent collaborative game evolution strategy to address issues of distributed resource sharing among virtual power plants, which is based on multicriteria matching and continuous learning. First, a peer matching mechanism is designed considering heterogeneous factor fusion and reasoning, which can enhance the adaptability and effectiveness of matching work. Then, a virtual–real hybrid game evolution framework is proposed based on continuous learning, in which a data-driven SE estimation method is developed to transform the traditional iterative response process into an efficient strategy estimation, thereby improving the efficiency of multiagent games. Furthermore, to further mitigate catastrophic forgetting in continuous learning, a multitimescale forward-feedback knowledge replay method is proposed to dynamically adjust replay timing and content, which can achieve knowledge consolidation through a combination of preventive maintenance and targeted repair, enhancing model evolution capabilities. Case studies validate that the proposed strategy exhibits superior performance in matching rationality, game convergence speed, and resistance to knowledge forgetting.

The success of deep learning models heavily depends on high-quality labeled data, yet noisy labels are inevitable in large-scale datasets. Existing methods often suffer from confirmation bias and overlook the informative value of hard but clean samples. To address these challenges, we propose Local Semantics Refinement of Adaptive Representations (LFDA), a novel framework that adaptively refines label quality by leveraging local feature consistency and representation alignment. LFDA introduces a Local Consistency Score module that evaluates the similarity among local samples in the latent space to distinguish clean from noisy labels. In addition, a confidence neighborhood is further constructed to provide local reference guidance, enabling more accurate identification and correction of noisy instances. To enhance semantic reliability, LFDA integrates a Reliability-Aware Representation Alignment (RRA) module that aligns high-confidence sample representations to implicitly refine low-confidence instances via soft supervision. Extensive experiments on both synthetic and real-world noisy datasets demonstrate that LFDA consistently outperforms state-of-the-art label noise learning methods. The results confirm its good robustness, generalization ability, and effectiveness in handling diverse and complex noise conditions.

The operation and control of active distribution networks (ADNs) are becoming increasingly important due to the high penetration of renewable energy (RE). The inherent uncertainty of RE can affect the stability and efficiency of ADN operations. To mitigate the inherent uncertainty and rapid variability of the high RE penetration in ADNs, this article uses an online ADN reconfiguration (ADNR) approach to ensure swift responses to RE fluctuations. Unlike traditional deep reinforcement learning (DRL)-based methods, which typically model the ADNR as a Markov decision process (MDP) and rely on historical ADN data to train the DRL agent, this approach may lead to a mismatch between the MDP's characteristics and the actual ADNR and pose challenges in handling scenarios that do not exist in the training data. To address this issue, this article proposes an online-offline DRL framework for online ADNR. Initially, during the offline stage, ADNR is formulated as a state-driven Markov decision process, which incorporates the operational characteristics of the ADN. Following this, a state-driven proximal policy optimization (SD-PPO) algorithm is proposed to enhance the generalization capability of DRL. In the subsequent step, we present the optimized action proximal policy optimization (OA-PPO) algorithm, which performs personalized training based on SD-PPO to further improve DRL performance in the online stage. The proposed approach is applied to three IEEE ADN systems. Numerical results demonstrate the effectiveness of our approach in reducing power loss and enhancing RE accommodation. Furthermore, detailed comparisons with other DRL and traditional ADNR algorithms confirm the superior computational performance of our proposed method.

This article considers the problem of output feedback-based decentralized periodic event-triggered control (PETC) for a class of large-scale systems with unknown nonlinear interconnections and measurement delays. When only the delayed sampled-data measurement output is accessible, a novel decentralized high-gain observer is first proposed based on an output predictor for each subsystem. Then, a set of decentralized sampled-data output feedback controllers that are driven by asynchronous periodic event-triggering conditions is developed to globally exponentially stabilize the large-scale systems. With the help of small-gain arguments and feedback domination approach, a rigorous stability analysis shows that there exist some sufficient conditions to ensure the global exponential stability of the overall systems. Different from the sample-and-hold implementation of output information, this article employs the prediction technique to obtain the current output prediction for each subsystem, which in turn effectively compensates for the undesirable effects of measurement delays and information loss. Finally, simulation results are presented to demonstrate the effectiveness of the proposed control method.

This article focuses on dynamic integral sliding mode control (SMC) of uncertain Takagi–Sugeno fuzzy delayed systems on time scales. SMC approaches for both continuous-and discrete-time fuzzy delayed systems are designed in a unified framework. First, we design a dynamic controller to guarantee global asymptotic stabilization (GAS) withH∞performance of the addressed systems. Second, to better adapt to the uncertainty characteristics of fuzzy models, an integral sliding mode surface (SMS) considering states, inputs, and uncertainties is proposed, which is an important contribution of this article. By utilizing the Lyapunov function and timescale calculus, it is shown that all states of the addressed control system can be driven close enough to the SMS and global asymptotic convergence (GAC) of the sliding motion can be ensured under matrix inequality criteria. In addition, the chattering phenomenon near the origin of discrete-time SMC system can be avoided in this article. Finally, three simulation examples are offered to illustrate the feasibility of the proposed control schemes.

This article proposes a novel distributed neuro-adaptive 3-D formation tracking control framework of multiple autonomous underwater vehicles (multi-AUVs) subject to marine environmental disturbances. On the one hand, we assume that all AUVs can obtain the real-time states. By introducing a series of variable transformations, the multi-AUV system is transformed into an underactuated nonlinear system with virtual control input. Radial basis function neural networks (RBFNNs), whose weights are updated online, are utilized to approximate nonlinear functions. Considering environmental disturbances, a virtual controller is designed such that all AUVs track the leader while maintaining the desired formation geometry. Then, the actual controller is given as an adaptive form according to the virtual control signals. On the other hand, we assume that all AUVs can only obtain the sampling states of themselves and their neighbors under the predefined event-triggered conditions. Multi-AUV system is transformed into a second-order system with complex nonlinear dynamics, then their states are reconstructed via a neuro-adaptive state observer using sampling states, and a virtual controller is proposed such that all AUVs track the leader while maintaining the desired formation geometry under local communication with no Zeno behavior. Finally, numerical simulations are carried out to demonstrate the effectiveness of the proposed control design.

Extracting a small subset of crucial rationales from the full input is a key problem in explainability research. The most widely used fundamental criterion for rationale extraction is the maximum mutual information (MMI) criterion. In this paper, we first demonstrate that MMI suffers from diminishing marginal returns. Once part of the rationale has been identified, finding the remaining portions contributes only marginally to increasing the mutual information, making it difficult to use MMI to locate the rest. In contrast to MMI that aims to reproduce the prediction, we seek to identify the parts of the input that the network can actually utilize. This is achieved by comparing how different rationale candidates match the capability space of the weight matrix. The weight matrix of a neural network is typically low-rank, meaning that the linear combinations of its column vectors can only cover part of the directions in a high-dimensional space (high-dimension: the dimensions of an input vector). If an input is fully utilized by the network, it generally matches these directions (e.g., a portion of a hypersphere), resulting in a representation with a high norm. Conversely, if an input primarily falls outside (orthogonal to) these directions, its representation norm will approach zero, behaving like noise that the network cannot effectively utilize. Building on this, we propose using the norms of rationale candidates as an alternative objective to MMI. Through experiments on four text classification datasets and one graph classification dataset using three network architectures (GRUs, BERT, and GCN), we show that our method outperforms MMI and its improved variants in identifying better rationales. We also compare our method with a representative LLM (llama-3.1-8b-instruct) and find that our simple method gets comparable results to it and can sometimes even outperform it.

Recent advances in video action recognition have achieved remarkable performance in coarse-grained macro-action classification by leveraging large-scale visual backbones and transformer architectures. However, extending these successes to fine-grained micro-action recognition remains a fundamental challenge due to the subtlety, brevity, and low motion intensity of micro-actions. In this paper, we propose a high-capacity framework for micro-action recognition, enhancing both representation learning and decision robustness. We scale to large-scale backbones using the VideoMAEv2 Giant model, enabling the extraction of finer spatial-temporal features. A Temporal-Spatial Connector (TSC) is introduced to dynamically highlight discriminative temporal frames and spatial regions, strengthening the model's focus on subtle motion cues critical for micro-action identification. To stabilize optimization and fully exploit the capacity of large models, we design a four-phase progressive training strategy, encompassing linear probing, full fine-tuning, connector-specific optimization, and classifier head refinement. Furthermore, we propose a novel ensemble decision mechanism that integrates Top-K predictions from diverse models via a Large Language Model (LLM), enhancing prediction consistency and robustness through multimodel consensus. Our method achieves an F1mean of 76.54% on the MA-52 dataset, ranking 3rd in the 2025 Micro-Action Analysis Grand Challenge and advancing the state of the art in fine-grained video understanding.

RNA N$^{6}$-methyladenosine (m$^{6}$A) is a critical epigenetic modification closely related to rice growth, development, and stress response. m$^{6}$A accurate identification, directly related to precision rice breeding and improvement, is fundamental to revealing phenotype regulatory and molecular mechanisms. Faced on rice m$^{6}$A variable-length sequence, to input into the model, the maximum length padding and label encoding usually adapt to obtain the max-length padded sequence for prediction. Although this can retain complete sequence information, resulting in sparse information and invalid padding, reducing feature extraction accuracy. Simultaneously, existing rice-specific m$^{6}$A prediction methods are still at an early stage. To address these issues, we develop a new end-to-end deep learning framework, MFDm$^{6}$ARice, for predicting rice m$^{6}$A sites. In particular, to alleviate sparseness, we construct a multi-kernel feature fusion module to mine essential information in max-length padded sequences by multi-kernel feature extraction function and effectively transfer information through global-local dynamic fusion function. Concurrently, considering the complexity and computational efficiency of high-dimensional features caused by invalid padding, we design a downsampling residual feature embedding module to optimize feature space compression and achieve accurate feature expression and efficient computational performance. Experiments show that MFDm$^{6}$ARice outperforms comparison methods in cross-validation, same- and cross-species independent test sets, demonstrating good robustness and generalization. The application on maize m$^{6}$A indicates the MFDm$^{6}$ARice's scalability. Further investigations have shown that combining different kernel features, focusing on global channel-local spatial, and employing reasonable downsampling and residual connections can improve feature representation and extraction, ensure effective information transfer, and significantly enhance model performance.

Multimodal emotion recognition in conversation (MERC) has garnered substantial research attention recently. Existing MERC methods face several challenges: (1) they fail to fully harness direct inter-modal cues, possibly leading to less-than-thorough cross-modal modeling; (2) they concurrently extract information from the same and different modalities at each network layer, potentially triggering conflicts from the fusion of multi-source data; (3) they lack the agility required to detect dynamic sentimental changes, perhaps resulting in inaccurate classification of utterances with abrupt sentiment shifts. To address these issues, a novel approach named GraphSmile is proposed for tracking intricate emotional cues in multimodal dialogues. GraphSmile comprises two key components, i.e., GSF and SDP modules. GSF ingeniously leverages graph structures to alternately assimilate inter-modal and intra-modal emotional dependencies layer by layer, adequately capturing cross-modal cues while effectively circumventing fusion conflicts. SDP is an auxiliary task to explicitly delineate the sentiment dynamics between utterances, promoting the model's ability to distinguish sentimental discrepancies. GraphSmile is effortlessly applied to multimodal sentiment analysis in conversation (MSAC), thus enabling simultaneous execution of MERC and MSAC tasks. Empirical results on multiple benchmarks demonstrate that GraphSmile can handle complex emotional and sentimental patterns, significantly outperforming baseline models.

Black-Box Knowledge Distillation (B2KD) is a conservative task in cloud-to-edge model compression, emphasizing the protection of data privacy and model copyrights on both the cloud and edge. With invisible data and models hosted on the server, B2KD aims to utilize only the API queries of the teacher model's inference results in the cloud to effectively distill a lightweight student model deployed on edge devices. B2KD faces challenges such as limited Internet exchange and edge-cloud disparity in data distribution. To address these issues, we theoretically provide a new optimization direction from logits to cell boundary, different from direct logits alignment, and formalize a workflow comprising deprivatization, distillation, and adaptation at test time. Guided by this, we propose a method, Mapping-Emulation KD (MEKD), to enhance the robust prediction and anti-interference capabilities of the student model on edge devices for any unknown data distribution in real-world scenarios. Our method does not differentiate between treating soft or hard responses and consists of: 1) deprivatization: emulating the inverse mapping of the teacher function with a generator, 2) distillation: aligning low-dimensional logits of the teacher and student models by reducing the distance of high-dimensional image points, and 3) adaptation: correcting the student's online prediction bias through a graph propagation-based only-forward test-time adaptation algorithm. Our method demonstrates inspiring performance for edge model distillation and adaptation across different teacher-student pairs. We validate the effectiveness of our method on multiple image recognition benchmarks and various Deep Neural Network models, achieving state-of-the-art performance and showcasing its practical value in remote sensing image recognition applications.

Out-of-distribution (OOD) detection presents a significant challenge in deploying pattern recognition and machine learning models, as they frequently fail to generalize to data from unseen distributions. Recent advancements in vision-language models (VLMs), particularly CLIP, have demonstrated promising results in OOD detection through their rich multimodal representations. However, current CLIP-based OOD detection methods predominantly rely on single-modality in-distribution (ID) data (e.g., textual cues), overlooking the valuable information contained in ID visual cues. In this work, we demonstrate that incorporating ID visual information is crucial for unlocking CLIP's full potential in OOD detection. We propose a novel approach, Dual-Pattern Matching (DPM), which effectively adapts CLIP for OOD detection by jointly exploiting both textual and visual ID patterns. Specifically, DPM refines visual and textual features through the proposed Domain-Specific Feature Aggregation (DSFA) and Prompt Enhancement (PE) modules. Subsequently, DPM stores class-wise textual features as textual patterns and aggregates ID visual features as visual patterns. During inference, DPM calculates similarity scores relative to both patterns to identify OOD data. Furthermore, we enhance DPM with lightweight adaptation mechanisms to further boost OOD detection performance. Comprehensive experiments demonstrate that DPM surpasses state-of-the-art methods on multiple benchmarks, highlighting the effectiveness of leveraging multimodal information for OOD detection. The proposed dual-pattern approach provides a simple yet robust framework for leveraging vision-language representations in OOD detection tasks.

Robust optimization (RO) has been widely used in the hydrogen-data-center microgrid (H2-DCMG) optimal operations. However, the operation results based on RO are too conservative. Statistical feasibility can be introduced into RO to reduce conservatism. Therefore, statistical feasibility-based RO is adopted in the H2-DCMG operations optimization. In this study, a data-driven statistical-feasibility-based robust rolling optimization framework is constructed for the optimal operations of the c-DCMG. In this framework, DC temperature is influenced by the uncertain outside temperature, and an uncertainty set is needed to express the uncertain outside temperature. Unfortunately, the existing studies that construct the uncertainty set satisfying statistical feasibility only constructs the ellipsoid uncertainty sets. The ellipsoid uncertainty sets will be converted into the second-order cone constraints, which will increase the complexity when solving. In this study, the Statistical-Guarantee-based Vertex Link (SGVL) algorithm is proposed to construct the polyhedron uncertainty sets, which are used to describe the uncertainty of the outside temperature. Moreover, statistical feasibility-based DC temperature bound is guaranteed by the optimal operations obtained based on these uncertainty sets. Compared with the traditional ellipsoid uncertainty set, the polyhedron uncertainty set can reduce the complexity of the optimization problem and improve the efficiency of the solution process. Case studies based on the real-world temperature dataset are processed. The results show that the introduction of statistical feasibility can reduce the total operation cost by 0.29%~0.64%. The average solving time of the optimization problems based on the polyhedron uncertainty sets constructed using the SGVL algorithm also reduces by 7%~13%. The cases also verify that other uncertainty parameters and different kinds of forecasters do not influence the performance and effectiveness of the framework. Note to Practitioners—This paper focuses on optimizing the operations of a hydrogen-data center microgrid (H2-DCMG) to minimize its operation cost in the way of statistical feasibility-based robust optimization. The statistical feasibility is introduced into the framework for the relaxation of the data center (DC) temperature bound. The Statistical-Guarantee-based Vertex Link (SGVL) algorithm is proposed to construct the uncertainty set of outside temperature that guarantees the dispatch results satisfying statistical feasibility. In practice, the settings of the framework and the data used in the SGVL are significant. Firstly, in the rolling dispatch, the length of the forecasting period needs also to be set after the trade-off between the solving efficiency, the dispatch economy and the DC temperature stability. Secondly, the pre-set DC temperature bound is advisable to be set below the maximum temperature that does not harm its normal operation. When constructing the uncertainty sets, the two parameters of the statistical feasibility need also to be set after the trade-off between the economics of dispatch results and the acceptability of constraint violations. The dataset and the actual forecasting error need to remain equally distributed to guarantee the performance of the dispatch of the proposed framework. The proposed framework can be readily implemented and integrated into the actual dispatch of H2-DCMG.

The accurate characterization of net load uncertainty can ensure the economy and stability of renewable power systems operation. In existing studies, Gaussian mixture model (GMM) and Dirichlet process mixture model (DPMM) are powerful tools for characterizing the uncertainty of monthly net load. However, the correlation among the Gaussian components and the common distribution characteristics among the monthly net load are not considered in these studies, which may lead to lower characterization accuracy. To solve these issues, we propose a hierarchical uncertainty characterization method for monthly net load considering Gaussian component reduction and correlation. Firstly, the method obtains the Gaussian mixture components and the label information of the annual net load by the Expectation-Maximum (EM) algorithm. Then, the Gaussian components of the monthly net load are corrected by the label information, which helps extract the common distribution characteristics among the monthly net load. On this basis, the corrected Gaussian components are considered as the base distribution. A component-preserving Expectation-Maximum (CPEM) algorithm is developed for component reduction. This realizes the uncertainty characterization of the monthly net load with high accuracy and lower time consumption. Importantly, the temporal correlation of net load is converted into the correlation among the Gaussian components, which are explicitly characterized by the Spearman coefficient. Finally, the superiority of the proposed method is verified with actual data collected in Australia. Note to Practitioners—In this article, we address the issue of monthly net load uncertainty characterization in renewable power systems. A hierarchical uncertainty characterization method is proposed to enable the distribution characteristic extraction of monthly net load. Most existing works on uncertainty characterization use GMM and DPMM, where the internal information in the net load data (e.g., distribution and component correlation information) is poorly utilized. This poses a considerable challenge for accurately analyzing the net load uncertainty. To this end, the article proposes a hierarchical uncertainty characterization method for monthly net load considering Gaussian component reduction and correlation. In this method, the common and typical distribution characteristics of monthly net load are considered by the CPEM algorithm. Moreover, the temporal information of the net load is transformed as the correlation among the Gaussian components, which are integrated into the E-step in the proposed CPEM algorithm. Since more internal information is utilized, the proposed method is helpful in characterizing the net load uncertainty by means of high accuracy and lower time consumption. The results of the uncertainty characterization can be readily implemented in the planning operation and real-time dispatch of renewable power systems.

This work proposes a bio-inspired decision-making memristive circuit drawing on Hull’s secondary learning system. This circuit can not only mimic the decision-making initiated by secondary drive stimuli and shaped and guided by secondary reinforcers via integrating classical conditioning (CC) and operant conditioning (OC), but also consider the factors that influence decision-making, such as demand states, incentive motivation, and habit strength. These bionic functions have not yet been implemented by existing memristive circuits. Our circuit primarily includes CC module, drive regulation module, habit memory module, incentive generation module, and winner-takes-all module, which is designed through a modular hierarchical circuit design method. Memristors play a core role in our circuit and enable the circuit to perform brain-like online learning in an in-memory computing way, which has power and area advantages. The PSPICE-based simulations in various scenarios show that our circuit has a strong adaptive decision-making ability since more bionic features are considered. The proposed circuit can be applied to a bionic intelligent robot, enabling the robot capable of autonomous associative learning abilities to perform complex tasks such as detection and rescue.Note to Practitioners—This work is motivated by the problem of neuromorphic circuit design for bio-inspired learning and decision-making. To realize brain-like online in-situ learning in an in-memory computing way and enhance the adaptability of the circuit in dynamic environment, a memristive circuit integrating CC and OC is proposed. Referring to Hull’s secondary learning system, the proposed circuit takes into account factors that affect decision-making, such as demand states, incentive motivation, and habit strength, as well as the fact that decision-making processes can be evoked by secondary drive stimuli and shaped by secondary reinforcers, which is unaddressed by existing memristor-based works. The bionic foraging simulations in various scenarios show that our circuit can make favorable adaptive decisions based on various cues categorized by associative memories when engaging with their environment. Such a bio-inspired memristive circuit system can be applied to bionic robots or rescue detection robots through large-scale integration to achieve adaptive learning and decision-making of complex tasks with low power consumption.

Driver fatigue is a critical factor that lead to traffic accidents with a high fatality rate. Electroencephalogram (EEG) is one of the most reliable indicators to objectively assess fatigue status, but recognizing fatigue driving status from it is still an essential and challenging problem. In this paper, we propose a multiscale global prompt Transformer (MsGPT) deep learning model, which can automatically recognize driver fatigue end-to-end. First, we construct an intra-inter-scale cascade framework based on Transformer with a multiscale convolutional patch embedding (MC-PatchEmbed), and guide global-local feature interaction by adding a global prompt token throughout. Second, to efficiently integrate intra-scale and inter-scale feature information, we design a mixed token by aggregating the output from the intra-scale, which includes rich low-level feature information for multiscale. Moreover, a novel learnable query is introduced into multi-head self-attention (MSA) to reduce the computational complexity to linear level. Experiments are conducted on the SEED-VIG dataset and the SADT dataset with both intra-subject and inter-subject settings to evaluate the performance of MsGPT, and the results show that MsGPT greatly outperforms various methods in terms of the classification evaluation metrics of EEG-based fatigue driving.Note to Practitioners—This paper considers the use of raw EEG data to recognize the driver fatigue state. Existing methods mainly rely on manually extracted EEG features and convolutional neural network (CNN) based inference. However, the large intra-individual and inter-individual differences greatly limit the extraction of EEG fatigue features. This paper suggests a multiscale global prompt Transformer (MsGPT) deep learning model. This model leverages a shared weighting mechanism to construct an inter-to intra-scale multiscale framework that can capture refined fatigue features not achievable at a single scale, we incorporate a new Transformer of the global prompt mechanism, which facilitates multiscale local-to-global fusion of long-term physiological signals. Our experiments demonstrate the superiority of our method on two datasets with intra-subject and inter-subject settings. The proposed method can be readily deployed in the automatic driving assistance system to alert drivers to avoid or reduce traffic accidents caused by excessive fatigue.

Memristor crossbar array (MCA) is a computing-in-memory (CIM) module for computational acceleration. However, conventional read-write (R-W) circuits for MCA rely heavily on external components and have shortcoming in long writing times. To address these issues, we propose a parallel R-W circuit with a fast amplitude-adaptive matching (AAM) scheme. The fast AAM scheme is designed to accelerate time of writing memristors in MCA by adaptively adjusting the writing voltage to match an optimal amplitude. The architecture of parallel R-W circuit are outlined with the implementation of R-W processes. The design principles of parallel R-W unit, fast AAM unit, and associated control logic are described for analyzing their functions within the circuit. By integrating the parallel R-W circuit with MCA, a computing block is leveraged in a neural network for classification tasks. The experimental results demonstrate that the neural network with multiple computing blocks maintains high accuracy across various datasets (over 80%). Furthermore, the proposed AAM scheme achieves an average 54% reduction in writing time compared to that one without AAM scheme. Compared to the pulse-width scheme, the parallel R-W circuit exhibits an average$3.05\times $speedup in adjusting a$5\times 5$MCA, and the speedup efficiency is higher as the increases of MCA size.

The capability of unisensory and multisensory information processing is crucial for bio-inspired intelligent systems. Based on biological nonassociative learning (NAL) and multisensory integration (MSI) mechanisms, a bio-inspired memristor-based neuromorphic circuit is proposed, which bridges multiple unisensory channels with a multisensory mutual associative memory (MMAM) unit. Inspired by the gill-withdrawal reflex in Aplysia, unisensory channels are capable of NAL processes, including habituation, sensitization, dishabituation, and spontaneous recovery, providing adaptation to innocuous stimuli and sensitivity to noxious stimuli. In light of the response enhancement and depression in the superior colliculus under multisensory cues, the MMAM unit enables the interaction between multiple sensory stimuli, thereby facilitating multisensory enhancement and depression, collectively known as MSI. By leveraging the proposed circuit, the artificial nociceptor and semantic satiation are mimicked. Furthermore, circuit performance analyses demonstrate the robustness and device tolerance. By further incorporating visual, auditory, tactile, olfactory, and gustatory sensors and scaling up the circuit, the neuromorphic system is promising for intelligent robot platforms with enhanced environmental perception and cognition capabilities.

In this paper, the fixed-time intra/inter-layer output synchronization problem of output-coupled multiplex networks is investigated utilizing a dynamic event-triggered control method. Firstly, to solve the issue of unavailability of node states resulting from uncontrollable factors, a multiplex networks model with observable intra/inter-layer output coupling information is constructed. Subsequently, two dynamic event-triggered control strategies based on output information are proposed, on the basis of which the fixed-time output synchronization criteria are established and Zeno behavior is excluded. The controllers proposed in this paper replace the common linear terms and multiple power-law terms in the existing fixed-time controllers with an exponential term based on the output errors, and also no longer include the intra/inter-layer coupling information of the nodes, making the form of the controllers more streamlined and easier to implement the control strategies. Finally, the effectiveness of the designed control protocols is verified by some numerical simulations based on Chua’s circuit as well as spacecraft formation control.

This paper focuses on addressing the practical finite-time synchronization (PFTS) problem of heterogeneous fractional-order complex dynamical networks (FCDNs) through event-triggered feedback control (ETFC). Firstly, a novel practical finite-time stability lemma is proposed based on the fractional-order differential inequality$_{t_{0}}^{C}D_{t}^{\alpha } V\left ({{ t }}\right) \le - {p_{1}}V\left ({{ t }}\right) - {p_{2}}{V^{\beta } }\left ({{ t }}\right) + q$, which plays a crucial role in analyzing PFTS. Secondly, a novel ETFC protocol is designed where the information transmission of the controller occurs at a sequence of state-dependent instants. Thirdly, using the aforementioned lemma and fractional Lyapunov theory, synchronization criteria for heterogeneous FCDNs can be derived, and Zeno behavior is excluded. Finally, the numerical example involving the PFTS of a fractional-order Lorenz’s circuit is provided to demonstrate the effectiveness of the proposed theoretical results.

Currently, studies on memristor-based operant conditioning circuits concentrate on the learning process of the single behaviors, without attention to the chaining process composed of multiple behaviors. This paper proposes a memristor-based circuit design of biological behavior chain, which consists of behavior modules, learning modules, delay modules, and satiety modules. After learning, the mouse can complete a chain process consisting of three behaviors: pressing button A to eliminate electrical stimulation, pressing button B to obtain food, and pressing button C to open the cage. Behavior and delay modules are used to simulate these three behaviors, the learning module is used to simulate the learning process of chaining between two behaviors and chaining between three behaviors. In addition, the study also explored the effects of mouse satiety on the experiment, as well as processes such as natural forgetting and relearning. The simulation results in PSPICE indicate that the circuit is capable of effectively simulating the aforementioned functions. Furthermore, Monte Carlo analysis and temperature simulation analysis are conducted on the circuit. The simulation results confirm that the circuit exhibits good stability.

Emotional support conversation (ESC) aims to alleviate the negative emotions of help-seekers by providing psychological assistance. Existing approaches typically overlook the abundant annotations contained in the ESC dataset, such as the situation descriptions and feedback scores of seekers, which limits their performance. In an effort to utilize the annotation information to enhance the emotional support ability of the backbone, we propose a three-stage training method called BlenderBot-ThTra for ESC systems. The proposed BlenderBot-ThTra involves the following three training processes: fine-tuning with supplemental feedback utterance, fine-tuning with auxiliary situation restoration, and calibration with the helpfulness estimation. The first stage aims to intensify the backbone's perception of conversational context, the second stage propels the backbone into excavating the causes of the emotional distress faced by the seeker. In the third stage, we leverage a Bayesian method based on the seeker's feedback scores to train a helpfulness evaluation model, then exploit a contrastive learning method to calibrate the ESC backbone. We conduct experiments on the standard multiturn ESC dataset, and the results demonstrate that BlenderBot-ThTra has a significant advantage in generating more supportive and adaptive responses.

Owing to their ability to effectively characterize the memory effect of magnetic flux, specifically in relation to the effect of external electromagnetic radiation, memristors have elicited widespread interest in the construction of neural networks with complex dynamics. This work proposes a novel memristive multiscroll multistable neural network (MMSMSNN), wherein multistable threshold memristors are used to describe external electromagnetic radiation effects. Numerical simulations show that the MMSMSNN can yield any number of cubic lattice multiscroll attractors by adjusting the internal parameters of memristors. Another highlight is that it can also be able to yield abundant initial offset boosting behaviors, i.e., different kinds of infinitely many homogeneous coexisting attractors, including linearly arranged homogeneous coexisting attractors, planar lattice-distributed homogeneous coexisting attractors, and cubic lattice-distributed homogeneous coexisting attractors. In addition, hardware experiments based on the CH32V307 microcontroller are carried out to demonstrate the numerical findings. Finally, a new secure medical image communication scheme is designed to investigate the MMSMSNN in practical applications, and performance analyses reveal its superiority and high security.

In this article, the event-triggered finite-time stabilization of time-scale delayed Takagi-Sugeno (T-S) fuzzy systems is studied. By comparing strategies, inequality techniques, and time scale theory, finite-time stabilization criteria for the systems are derived that do not require differentiability of the time delay, and the controller is designed in a simple form that does not rely on power functions or delayed state feedback controllers. Corresponding results cover both continuous-time and discrete-time cases, and construct a unified theoretical framework for the finite-time analysis of the time-scale delayed systems. Meanwhile, the proposed event-triggered mechanism can avoid Zeno behavior and reduce the consumption of communication resources. The validity of the theoretical results is verified by two simulation experiments.

This study investigates the semi-global fixed-time stability (SGFTS) and global fixed-time stability (GFTS) of nonlinear impulsive systems (NISs). A key challenge in analyzing the SGFTS of such systems lies in the evolving integration methods caused by the impulses. To address this, we dynamically partition the semi-global attraction set (SGAS) and solve the corresponding differential equations within each subset. Additionally, by constructing the transition dynamics of impulse points and iteratively computing these points, we establish the conditions for SGFTS under both stabilizing and destabilizing impulses. For GFTS, the primary difficulty arises from the distinct trajectories and dynamics of points located inside and outside the SGAS. To overcome this, we introduce the concept of the maximum-minimum impulse interval and derive a sufficient condition that ensures the system can enter the SGAS from a distance under a finite number of impulses. Furthermore, we develop a criterion for GFTS under varying impulse degrees and provide convergence time estimation based on the research on SGFTS of NIS. Finally, numerical examples are presented to validate the theoretical results. Notably, in Example 3, a fixed-time impulse controller is designed based on the proposed theoretical framework to achieve global stabilization of complex systems. This example highlights the potential applications of this work in the field of control.

In this article, leader-following consensus of time-scale-type heterogeneous nonlinear multiagent systems (HNMASs) is investigated with dynamic periodic event-triggered mechanism (DPETM). The event detection period in DPETM is determined by a function-dependent threshold, whose initial value and the value at each periodic event detection instant are used for the update of an auxiliary function in the DPETM. Furthermore, the auxiliary function with periodic jumps serves as a detection threshold. To guarantee the nonincreasing behavior of the designed non-negative analysis function, a weighted function is devised that shares the same derivative form as the function that determines the detection period during each detection period. Then, by integrating the theory of time scales and graph theory, leader-following consensus is achieved in a periodic communication fashion with fewer sampling updates. Two examples are presented to illustrate the validity of the results.

In this article, we propose a fuzzy adaptive network (FAN) with learnable parameters for mixed-integer optimization. Specifically, by leveraging a recurrent network to infer the discretization parameters, the FAN is implemented in an easy-to-implement discrete-time format. FAN possesses the dynamic behavior of a high-precision numerical differential rule and maintains a simple network structure. In addition, a fuzzy mechanism is incorporated to adjust the step size. Sufficient conditions are derived such that the proposed FAN is globally exponentially convergent to a Karush–Kuhn–Tucker point. In the presence of nonconvexity in objective functions or constraints, multiple FANs operate concurrently in a hybrid intelligent algorithm. Finally, multiple comparative experiments are conducted to demonstrate the superiority of the proposed FAN in terms of time efficiency and solution quality.

This article develops a methodology employing partial differential equations (PDEs) to facilitate the exponential deployment of large-scale heterogeneous nonlinear multiagent systems (MASs). The considered MASs comprise a multitude of nonlinear first-order agents (FOAs) and second-order agents (SOAs). Two heterogeneous nonlinear PDEs are established to model the considered MASs by designing appropriate network communication protocols. Unlike previous PDE-based approaches for multiagent deployment, the topological weights between neighboring agents are defined as series-dependent. An informed agent, which is able to measure the location information of other agents and transmit its location information to neighboring agents through the communication network, is placed between the final FOA and the initial SOA. This novel network-based control scheme is referred to as single-point control, which could ensure the well-posedness and exponential stability of the error system. Accordingly, pointwise and distributed measurements are employed for delay-free and time-delayed cases, respectively. Numerical examples are provided in 3-D space to substantiate the obtained theoretical results.

Carbon trading mechanisms can constrain carbon emissions in energy supply, while demand response (DR) mechanisms have the potential to achieve low-carbon oriented energy consumption. Based on these, this paper focuses on mechanism-driven deep decarbonization methods of microgrid clusters (MGCs) from both source and load sides, and the key contributions are threefold. First, a novel dynamic vertical peer-to-peer carbon trading framework is proposed with profitability and fairness, in which a dynamic internal carbon emission right (CER) pricing mechanism is designed based on the supply–demand ratio of CERs and trading prices of the external primary carbon market. Then, a multiple structural DR mechanism is developed to increase the on-site utilization of renewable energy and decrease carbon emissions of MGCs, which considers the influence of electricity prices, the matching degree between renewable energy and loads, and these psychological preferences of users. Moreover, a DR feedback correction strategy is proposed to address the volatility of DR results caused by changes in load elasticity. Numerical results show that the proposed method can reduce carbon emissions by more than 15% without increasing the economic burden, which validates the effectiveness in improving the economic and low-carbon performance of MGCs' operation.

In the scenario of class-incremental learning (CIL), deep neural networks have to adapt their model parameters to non-stationary data distributions, e.g., the emergence of new classes over time. To mitigate the catastrophic forgetting phenomenon, typical CIL methods either cumulatively store exemplars of old classes for retraining model parameters from scratch or progressively expand model size as new classes arrive, which, however, compromises their practical value due to little attention paid toparameter efficiency. In this paper, we contribute a novel solution, effective control of the parameters of a well-trained model, by the synergy between two complementary learning subnetworks. Specifically, we integrate one plastic feature extractor and one analytical feed-forward classifier into a unified framework amenable to streaming data. In each CIL session, it achieves non-overwritten parameter updates in a cost-effective manner, neither revisiting old task data nor extending previously learned networks; Instead, it accommodates new tasks by attaching a tiny set of declarative parameters to its backbone, in which only one matrix per task or one vector per class is kept for knowledge retention. Experimental results on a variety of task sequences demonstrate that our method achieves competitive results against state-of-the-art CIL approaches, especially in accuracy gain, knowledge transfer, training efficiency, and task-order robustness. Furthermore, a graceful forgetting implementation on previously learned trivial tasks is empirically investigated to make its non-growing backbone (i.e., a model with limited network capacity) suffice to train on more incoming tasks.

The rapid development of renewable energy sources (RESs) has led to their increased integration into microgrids (MGs), emphasizing the need for safe and efficient energy management in MG operations. We investigate the methods of MG energy management, primarily categorized into model-based and model-free approaches. Due to a lack of incremental knowledge, model-based methods need to be reengineered for new scenarios during the optimization process, leading to reduced computational efficiency. In contrast, model-free methods can obtain incremental knowledge via trial-and-error in the training phase, and output energy management scheme rapidly. However, ensuring the safety of the scheme during the training phases poses significant challenges. To address these challenges, we propose a safe reinforcement learning (SRL) framework. The proposed SRL framework initially includes a safety assessment optimization model (SAOM) to evaluate scheme constraints and refine unsafe schemes for ensuring MG safety. Subsequently, based on SAOM, the MG energy management issue is formulated as an assess-based constrained Markov decision process (A-CMDP), enabling the SRL can be adopted in this issue. After that, we adopt a Lyapunov-based safety policy optimization for agent policy learning to ensure that policy updates are confined within a safe boundary, theoretically ensuring the safety of the MG throughout the learning process. Numerical studies highlight the superior performance of our proposed method. Specifically, the SRL framework effectively learns energy management policy, ensures MG safety, and demonstrates outstanding outcomes in the economic operation of MG.

Privacy of user is becoming increasingly significant in constructing efficient multiagent energy management systems for multimicrogrid (MMG). As an emerging privacy-protection method, federated learning (FL) has been used to prevent data breaches in the MMG-related field. However, with the ever-growing participants, the underlying communication burden existing in FL is evident. Besides, since the neural network layers collectively determine an agent's performance, the possible difference in layer convergence speeds would cause the inconsistency problem, that is, the FL may degrade the convergence rate of those fast-convergent layers, which weakens the overall performance of the agent. To address these issues, a communication-efficient FL (CEFL) algorithm is proposed in this study. Considering the cooperative relationship among layers, a layer evaluation (LE) mechanism is developed in CEFL to evaluate layer contribution through the Shapley value (SV), a profit distribution approach for coalitions. In this way, only partial layers with the highest contributions are selected to be uploaded to the server. In addition, instead of average parameters aggregation, a communication-efficient parameter aggregation method is proposed in CEFL to update the parameters of the global model (GM), in which an aggregation model (AM) is developed to receive parameters for aggregation. The performance of the proposed CEFL is verified by the numerical analysis of MMGs with 3-8 MGs participating. Furthermore, experiments investigate the influence of the hyperparameter in the CEFL and also demonstrate performance improvements, compared with the other four state-of-the-art algorithms.

To improve the acceleration performance, a hybrid state-triggered discretization (HSTD) is proposed for the adaptive gradient neural network (AGNN) for solving time-dependent linear equations (TDLEs). Unlike the existing approaches that use an activation function or a time-varying coefficient for acceleration, the proposed HSTD is uniquely designed from a control theory perspective. It comprises two essential components: adaptive sampling interval state-triggered discretization (ASISTD) and adaptive coefficient state-triggered discretization (ACSTD). The former addresses the gap in acceleration methods related to the variable sampling period, while the latter considers the underlying evolutionary dynamics of the Lyapunov function to determine coefficients greedily. Finally, compared with commonly used discretization methods, the acceleration performance and computational advantages of the proposed HSTD are substantiated by the numerical simulations and applications to robotics.

The trust-region method and projection neural networks are two branches of optimization approaches with different operational principles and characteristics. In this article, a trust-region projection neural network (TRPNN) is proposed by integrating the trust-region method and projection neural networks. TRPNN is a discrete-time neurodynamic optimization model that inherits the exploration-exploitation capability of the trust-region method and the local search capability of projection neural networks. TRPNN is theoretically proven to be convergent to a Karush-Kuhn-Tuchker (KKT) point of nonlinear programming problems. The efficacy of TRPNNs leveraged in a collaborative neurodynamic framework is numerically demonstrated for global optimization in the presence of nonconvexity in objective functions or constraints.

In this article, an online reinforcement learning (RL) control method through value iteration (VI) is developed to solve the optimal cooperative control problem for the unknown linear discrete-time multiagent systems (MASs). On the one hand, an online learning scheme with evolving policies is proposed in order to guarantee the stability of the MASs under immature policies generated by VI. Inspired by the event-triggered mechanism, the stability criterion is designed as a trigger to filter the admissible control policies, which eliminates the need to establish a monotonic value function sequence. On the other hand, an acceleration mechanism for the MASs is presented such that the convergence rate of VI can be accelerated. The relationship between the selection of the relaxation factor and the accelerated convergence process is elaborated. Simple backpropagation (BP) neural networks (NNs) are applied for the implementation. Two classical examples are introduced and simulation results are provided in order to substantiate the validity of the designed method.

This article proposes new theoretical results on the multiple Mittag-Leffler stability of almost periodic solutions (APOs) for fractional-order delayed neural networks (FDNNs) with nonlinear and nonmonotonic activation functions. Profited from the superior geometrical construction of activation function, the considered FDNNs have multiple APOs with local Mittag-Leffler stability under given algebraic inequality conditions. To solve the algebraic inequality conditions, especially in high-dimensional cases, a distributed optimization (DOP) model and a corresponding neurodynamic solving approach are employed. The conclusions in this article generalize the multiple stability of integer- or fractional-order NNs. Besides, the consideration of the DOP approach can ameliorate the excessive consumption of computational resources when utilizing the LMI toolbox to deal with high-dimensional complex NNs. Finally, a simulation example is presented to confirm the accuracy of the theoretical conclusions obtained, and an experimental example of associative memories is shown.

Most of the existing event-triggered mechanisms (ETMs) were designed according to the difference between the quadratic form of measurement errors and the quadratic form of sampling states (or real-time states). In order to reduce the amount of data transmission and develop ETMs for continuous-time and discrete-time delayed nonlinear systems (NSs) simultaneously, this article investigates quasi-synchronization (QS) of NSs on time scales based on a novel ETM, which is designed according to the convergence rate instead of measurement errors of the addressed systems. First, a novel ETM is designed under known nonlinear dynamics, and it is demonstrated that QS with given convergence rate and error level can be achieved under matrix inequality criteria. Second, if the nonlinear functions are unknown, we adapt our ETM to handle this special case. Not only QS but also complete synchronization with given convergence rate can be achieved under the ETMs. If the constructed Lyapunov functions passes through 0, the designed ETM will keep it at the origin. In this case, finite-time synchronization is achieved. Third, under the designed ETMs, it is proved that Zeno behavior can be excluded. At last, four numerical simulations are presented to demonstrate the feasibility and the advantage of the designed ETMs in this article.

In recent years, adaptive drive-response synchronization (DRS) of two continuous-time delayed neural networks (NNs) has been investigated extensively. For two timescale-type NNs (TNNs), how to develop adaptive synchronization control schemes and demonstrate rigorously is still an open problem. This article concentrates on adaptive control design for synchronization of TNNs with unbounded time-varying delays. First, timescale-type Barbalat lemma and novel timescale-type inequality techniques are first proposed, which provides us practical methods to investigate timescale-type nonlinear systems. Second, using timescale-type calculus, novel timescale-type inequality, and timescale-type Barbalat lemma, we demonstrate that global asymptotic synchronization can be achieved via adaptive control under algebraic and matrix inequality criteria even if the time-varying delays are unbounded and nondifferentiable. Adaptive DRS is discussed for TNNs, which implies our control schemes are suitable for continuous-time NNs, their discrete-time counterparts, and any combination of them. Finally, numerical examples on TNNs and timescale-type chaotic Ikeda-like oscillator with unbounded time-varying delays are carried out to verify the adaptive control schemes.

Zero-shot learning casts light on lacking unseen class data by transferring knowledge from seen classes via a joint semantic space. However, the distributions of samples from seen and unseen classes are usually imbalanced. Many zero-shot learning methods fail to obtain satisfactory results in the generalized zero-shot learning task, where seen and unseen classes are all used for the test. Also, irregular structures of some classes may result in inappropriate mapping from visual features space to semantic attribute space. A novel generative mixup networks with semantic graph alignment is proposed in this article to mitigate such problems. To be specific, our model first attempts to synthesize samples conditioned with class-level semantic information as the prototype to recover the class-based feature distribution from the given semantic description. Second, the proposed model explores a mixup mechanism to augment training samples and improve the generalization ability of the model. Third, triplet gradient matching loss is developed to guarantee the class invariance to be more continuous in the latent space, and it can help the discriminator distinguish the real and fake samples. Finally, a similarity graph is constructed from semantic attributes to capture the intrinsic correlations and guides the feature generation process. Extensive experiments conducted on several zero-shot learning benchmarks from different tasks prove that the proposed model can achieve superior performance over the state-of-the-art generalized zero-shot learning.

This article investigates the voltage restoration problem of direct current (DC) microgrids (MGs) under stealth attacks. The fact is that secondary control requires information exchange through sparse communication networks, which makes cybersecurity the key to achieving voltage restoration. Compared with the attack signals launched by nonintelligent attackers, stealthy attack signals are more difficult to capture. Similarly, since MGs actually operate in a closed environment, malicious attackers have limited knowledge of the MG’s structural information. Therefore, we reformulate the problem to a two-person zero-sum game between the attacker and defender and then seek the optimal voltage restoration control strategy. In the process of solving the optimal problem, both robustness and convergence of parameter estimation are considered in this article to enhance the feasibility of adaptive dynamic programming schemes. Additionally, while ensuring the voltage restoration performance of DC MGs and considering the constraints of actual network communication, a novel dynamic event-triggered mechanism is designed to reduce information transmission pressure. The effectiveness of the established control scheme is verified through real-time testing equipment built on OPAL-RT and comparison results.

In this article, the bipartite consensus problem of multiagent systems on time scales under switching topologies is investigated. A dynamic event-triggered control strategy is designed to reduce the number of triggers. Sufficient condition to guarantee the consensus is obtained by constructing a non-negative function and combining with the theory of time scale calculus. In addition, it is proved through a categorical discussion that the entire triggering sequence determined by the switching topology and the triggering function does not display Zeno behavior. Lastly, to confirm that the theoretical results are feasible, a numerical simulation and an application to spacecraft formation flight are provided.

Superdiffusion refers to the faster diffusion process in a multiplex network compared to that in an individual network. In this work, we study how interlayer connectivity affects the diffusion performance of a multiplex network. Based on spectral graph theory, we explore the principles of superdiffusion in multiplex networks. We prove that in a duplex network with identical structures, superdiffusion cannot occur under one-to-one interlayer connections. In addition, we prove that the dissimilarity of the Fiedler vector significantly enhances the network superdiffusion performance, which can lead to superdiffusion when selecting nodes with differential eigenvector components in the Fiedler vector for interlayer connections. We also prove that the upper bound of network diffusion with interlayer crossing-connections is limited by the maximum difference of the eigenvector components in the Fiedler vector. Finally, we verify the effectiveness of the theoretical results by numerical analysis.

Pinning control provides an effective approach to controlling large-scale networks and conserving control resources. This article presents a solution to pinning synchronization in directed networks with a precise index that measures the pinning synchronization capability of directed networks, capturing full topological information about the networks. Building upon this index, the article utilizes matrix analysis tools, such as the non-negative matrix theory and strongly connected decomposition to analyze the impact of network structures and controller parameters on the network synchronizability. Specifically, the study investigates the influence of the in-degree of unpinned nodes, the difference between in-degrees and out-degrees of nodes, strong connectivity components, and the linear feedback control gains on the network synchronizability. Moreover, the article addresses the challenge of optimally selecting pinned nodes by using a graph partitioning algorithm and a greedy node selection algorithm, which can be applied to effectively select pinned nodes in a large-scale network. Extensive simulations on a range of real-world directed networks validate the efficiency of the proposed algorithms and demonstrate their superiority over seven baseline algorithms.

This article investigates the practical finite-time spatial deployment of a class of large-scale heterogeneous nonlinear multi-agent systems (MASs), for which a novel hybrid analysis methodology based on ordinary differential equations (ODEs) coupled with partial differential equations (PDEs) is proposed. The assumption is made that a portion of the agents is sparsely distributed in space, while the other portion is densely distributed. By designing appropriate network communication protocols (NCPs), the dynamics of MASs are represented by a hybrid model consisting of several ODEs and a PDE. Particularly, the network topological weights are specifically designed as semi-Markov switched to better align with real communication situations of MASs, while complying with inconsistent switching rules. Moreover, for delay-free and time-delayed cases, this article proposes two novel projection-based adaptive neural control schemes and obtains two design criteria of controller gains, such that the practical finite-time stability of the tracking error systems could be guaranteed. Finally, numerical examples are provided to illustrate the effectiveness of the developed approaches.

This article establishes a series of theoretical findings on the sampled-data-based event-triggered (SET) output-feedback consensus for a class of heterogeneous high-order multiagent systems (MASs) with mismatched parametric uncertainties. First, a distributed SET leader observer is devised for each agent to estimate the information of leader, and its convergence is rigorously proven based on matrix theory and analysis approaches. Second, a continuous state observer is developed using solely SET output signal influenced by sensor faults is proposed to avoid the nondifferentiability of the virtual controller, and the SET output-feedback control protocol is designed via the backstepping technique. It is proven that the semi-global output consensus problem can be addressed through the proposed controller. Different from the existing work, the consensus protocol presented in this article further saves communicational and computational resources, because the SET mechanisms are deployed to each channel of MASs which merely need to discretely monitor the event-triggered (ET) conditions at sampling instants and transmit the information at triggered time. Besides, the Zeno behavior can be trivially prevented. Finally, a simulation example is depicted to demonstrate the validity of the proposed theoretical results.

The vast majority of published event-triggered mechanisms (ETMs) are constructed based on measurement errors, which introduces a problem naturally that they are updated when the measurement errors exceed the thresholds although the current obtained sampling states can make systems converge well. With this problem in mind, we redesign ETMs for quasi-synchronization of T-S fuzzy neural networks (FNNs) with time delays on time scales. First, a novel ETM is designed for continuous-time FNNs with time-varying delays to achieve quasi-synchronization, with which synchronization errors is suppressed to globally exponentially converge to a ball. Second, we introduce the ETM for continuous-time FNNs to discrete-time FNNs, owing to the existence of discrete-time states, the Lypunov function of synchronization errors run over the exponentially decay curve, but it can be suppressed to evolve under another exponentially decay curve. Third, for FNNs on time scales with constant and time-varying delays, we estimate the forward jump operator of the Lyapunov functions and design ETMs to guarantee that the Lypunov functions evolve under the exponentially decay curves, so quasi-synchronization can be achieved. Last but not least, we prove that Zeno behavior will not happen and four numerical examples are introduced to verify the validity and the superiority of the proposed ETMs in reducing information transmission.

This article focuses on achieving the finite-time synchronization (FTS) for fractional complex dynamical networks (FCDNs) using hybrid impulsive control. Initially, a novel framework for local FTS is developed, building upon the relaxed inequality${}_{{t_{k}}}^{C}D_{t}^{{\alpha}}V( t ) \le \chi V( t ) - \eta$. To expand the attraction domain within the local FTS framework, a piecewise fractional-order differential inequality based on impulsive control systems is proposed. Subsequently, a new hybrid control strategy is designed by integrating a simple feedback controller with an impulsive controller involving a finite number of impulses, which can be accurately calculated using the proposed impulsive degree. Additionally, a set of local/global FTS criteria is formulated, and the settling time can be explicitly estimated. Lastly, an illustrative example is presented to demonstrate the effectiveness of the derived results.

Thanks to the biomimetic properties of synaptic plasticity, memristors are often utilized to mimic biological neuronal synapses. This article presents a new memristor synapse coupling (MSC) approach for producing multidirectional multidouble-scroll attractors. Through adopting flux-controlled hyperbolic memristor synapses to couple a Hopfield neural network, a novel multidirectional multidouble-scroll Hopfield neural network (MDMDSHNN) is constructed. Theoretical results and numerical calculations indicate that MDMDSHNN is capable of producing any desired amount of multidirectional multidouble-scroll attractors, including unidirectional (1-D), bidirectional (2-D), and three-directional (3-D) multidouble-scroll attractors. Furthermore, an infinite amount of initial offset-boosted coexisting multidouble-scroll chaotic attractors possessing identical shapes but different positions, i.e., homogeneous extreme multistability are also found via switching the memristor initial values. Furthermore, to validate the physical implementability and practicality of MDMDSHNN, the digital hardware platform is performed. Finally, to investigate MDMDSHNN in practical application, an image encryption scheme with superior security performance is given by employing the homogeneous multidouble-scroll chaotic sequences, further illustrating good superiority and effectiveness of the present MSC method.

Deep neural networks are susceptible to catastrophic forgetting when trained on sequential tasks. Various continual learning (CL) methods often rely on exemplar buffers or/and network expansion for balancing model stability and plasticity, which, however, compromises their practical value due to privacy and memory concerns. Instead, this paper considers a strict yet realistic setting, where the training data from previous tasks is unavailable and the model size remains relatively constant during sequential training. To achieve such desiderata, we propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. This is achieved by the synergy between two key components: HSIC-Bottleneck Orthogonalization (HBO) implements non-overwritten parameter updates mediated by Hilbert-Schmidt independence criterion in an orthogonal space and EquiAngular Embedding (EAE) enhances decision boundary adaptation between old and new tasks with predefined basis vectors. Extensive experiments demonstrate that our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.

Class-incremental learning (CIL) aims to train a model to learn new classes from non-stationary data streams without forgetting old ones. In this paper, we propose a new kind of connectionist model by tailoring neural unit dynamics that adapt the behavior of neural networks for CIL. In each training session, it introduces a supervisory mechanism to guide network expansion whose growth size is compactly commensurate with the intrinsic complexity of a newly arriving task. This constructs a near-minimal network while allowing the model to expand its capacity when cannot sufficiently hold new classes. At inference time, it automatically reactivates the required neural units to retrieve knowledge and leaves the remaining inactivated to prevent interference. We name our model AutoActivator, which is effective and scalable. To gain insights into the neural unit dynamics, we theoretically analyze the model’s convergence property via a universal approximation theorem on learning sequential mappings, which is under-explored in the CIL community. Experiments show that our method achieves strong CIL performance in rehearsal-free and minimal-expansion settings with different backbones.

Multimodal Emotion Recognition in Conversation (ERC) plays an influential role in the field of human-computer interaction and conversational robotics since it can motivate machines to provide empathetic services. Multimodal data modeling is an up-and-coming research area in recent years, which is inspired by human capability to integrate multiple senses. Several graph-based approaches claim to capture interactive information between modalities, but the heterogeneity of multimodal data makes these methods prohibit optimal solutions. In this work, we introduce a multimodal fusion approach named Graph and Attention based Two-stage Multi-source Information Fusion (GA2MIF) for emotion detection in conversation. Our proposed method circumvents the problem of taking heterogeneous graph as input to the model while eliminating complex redundant connections in the construction of graph. GA2MIF focuses on contextual modeling and cross-modal modeling through leveraging Multi-head Directed Graph ATtention networks (MDGATs) and Multi-head Pairwise Cross-modal ATtention networks (MPCATs), respectively. Extensive experiments on two public datasets (i.e., IEMOCAP and MELD) demonstrate that the proposed GA2MIF has the capacity to validly capture intra-modal long-range contextual information and inter-modal complementary information, as well as outperforms the prevalent State-Of-The-Art (SOTA) models by a remarkable margin.

Multimodal emotion recognition in conversation (ERC) has garnered growing attention from research communities in various fields. In this paper, we propose a Crossmodal Fusion Network with Emotion-Shift Awareness (CFNESA) for ERC. Extant approaches employ each modality equally without distinguishing the amount of emotional information in these modalities, rendering it hard to adequately extract complementary information from multimodal data. To cope with this problem, in CFN-ESA, we treat textual modality as the primary source of emotional information, while visual and acoustic modalities are taken as the secondary sources. Besides, most multimodal ERC models ignore emotion-shift information and overfocus on contextual information, leading to the failure of emotion recognition under emotion-shift scenario. We elaborate an emotion-shift module to address this challenge. CFNESA mainly consists of unimodal encoder (RUME), cross-modal encoder (ACME), and emotion-shift module (LESM). RUME is applied to extract conversation-level contextual emotional cues while pulling together data distributions between modalities; ACME is utilized to perform multimodal interaction centered on textual modality; LESM is used to model emotion shift and capture emotion-shift information, thereby guiding the learning of the main task. Experimental results demonstrate that CFN-ESA can effectively promote performance for ERC and remarkably outperform state-of-the-art models.

Reservoir computing (RC) contains two significant variants: wide RC and deep RC. The hybrid wide-deep architecture absorbs their strengths with a powerful parallel processing capability while enhancing the memory capacity of reservoirs. However, the fully analog RC circuit combining the two structures has yet to be proposed, mainly due to unmanageable hierarchical signal processing. Here we report a full-analog memristive RC circuit with a hybrid wide-deep architecture comprising an input module, mask module, reservoir module, and readout module. The input module can generate continuous voltages with temporal sequences. The mask module provides parallel mask processes, laying the foundation for implementing a wide RC structure. The reservoir module includes dynamic memristors and postprocessing circuits. Dynamic memristors can produce high-dimensional reservoir states, and postprocessing circuits allow memristive reservoir circuits to be cascaded to achieve a deep RC structure. The readout module mainly consists of a nonvolatile memristor crossbar and an analog integrator, enabling an efficient multiplication-and-accumulation operation. The simulation results in LTspice illustrate that the memory capacity of the proposed circuit is 91.6% higher than that of wide RC. Moreover, it can efficiently perform temporal tasks, obtaining a high accuracy of 98.99% in arrhythmia detection.

For large-scale heterogeneous nonlinear multi-agent systems (MASs) consisting of abundant first-order and second-order agents, this paper presents a novel framework based on partial differential equations (PDEs) to facilitate their practically finite-time deployment in 2D or 3D space. First, through designing appropriate network communication protocols (NCPs), a heterogeneous nonlinear PDE model composed of a heat equation and a damped wave equation is constructed to characterize the collective dynamics of considered heterogeneous nonlinear MASs. Second, a single-point control strategy and a double-boundary control strategy are proposed, which could not only ensure the well-posedness of the closed-loop heterogeneous PDEs but also enable the finite-time deployment of multi agents. Notably, to better align with real MASs and operating environment, the network topologies and controllers are designed to be semi-Markov switched, while adhering to multiple asynchronous switching rules. Third, with the designed NCPs and control schemes, several sufficient conditions are derived to guarantee the practically finite-time stability of error systems. Finally, two numerical examples and an application example are conducted to validate effectiveness and practicability of the developed approaches.

In this contribution, the problem of fixed-time stabilization in multi-weighted complex networks via the novel adaptive pinning nonchattering control based on the linear matrix inequality (LMI) method, as well as its application to image protection is addressed. Different from the traditional methods, a novel fixed-time stable form is proposed and the convergence time is estimated based on beta function. Next, utilizing the designed continuous adaptive control strategy, a sufficient LMI condition is presented to ensure the fixed-time stabilization of multi-weighted complex networks. Furthermore, the novel nonchattering adaptive pinning control protocol is given to guarantee the fixed-time stabilization of the system only by controlling a small number of nodes. Note that a scheme of how to select the number of control nodes is put forward accordingly. Finally, the effectiveness of the proposed method is verified by the actual financial model. Meanwhile, a number of encryption experiments are carried out based on three networks, and the mean and variance of the encryption performance are calculated to show the stability and robustness of image encryption different from the existing research works.

Discrete memristors have been employed in discrete maps for the purpose of chaos generation and regulation. In this paper, a novel universal model for discrete memristors is proposed to generate multi-attractors. The classical Hénon map and Rulkov neuron are chosen as two examples to verify the effectiveness of the proposed memristor. Coexisting homogeneous attractors are identified in the phase space by memristor-induced offset boosting. An arbitrarily desired number of coexisting attractors is extracted by the appropriate feedback strength of the memristor. What adds further interest to this case is that the amplitude is rescaled by a memristor-related parameter that works well over an infinite range. Number-related parameters are extracted to rescale the oscillation range of the chaotic signals. Moreover, CH32-based circuit implementation is built, which aligns with numerical simulation results. Finally, coexisting homogeneous chaotic signals are tested to explore their robust performance in the application of pseudo-random number generator.

This article studies the global asymptotic stability problem of multiple integrators subject to uncertain parameters and high-order nonlinear perturbation. By introducing a modified saturation function with positive tunable parameters, three types of modified nested saturated controllers with different parameter assignments are proposed to handle the uncertain integrators with high-order nonlinear perturbation. The specific mechanism of the modified nested saturated controllers with different parameter assignments, in which control parameters appear before states, saturation functions, or saturation levels, is comprehensively analyzed. With the help of a modified saturation function, the global convergence domain of each state can be kept invariable in the saturation reduction analysis. Meanwhile, a large saturation level is obtained by adjusting positive tunable parameters. Combining with derivatives calculation and M -matrix-based comparison theory, proposed saturated control schemes are utilized to handle the considered system. Finally, an active magnetic bearing (AMB) system is revisited as an uncertain mechanical system and used to verify the feasibility of the proposed control designs.

Pinning control has been attracting wide attention for the study of various complex networks for decades. This article explores grounded theory on the pinning synchronization of the emerging multiplex dynamical networks. The multiplex dynamical networks under study can describe many real-world scenarios, in which different layers have distinct individual dynamics of node. In this work, we build the bridge between multiplex structures and network dynamics by using the Lyapunov stability theory and the spectral graph theory. Furthermore, by analyzing spectral properties of the grounded super-Laplacian matrices, we set up several graph-based synchronization criteria for multiplex networks via pinning control. In addition, we overcome the difficulties induced by distinct node dynamics in different layers, and find that interlayer coupling strengths promote intralayer synchronization of multiplex networks. Finally, a collection of numerical simulations verifies the effectiveness of theoretical results.

For a class of 2-D spatial distributed parameter systems (DPSs) with space-dependent diffusivity, this article aims to achieve exponential realization of their desired profiles. To reduce the number of required sensors and actuators, a planar output feedback boundary control strategy is proposed with combining two nonfull-domain measurement methods, boundary collocated measurement and planar linear measurement, in which only two boundaries of the considered 2-D spatial DPSs are controlled and a little output information is measured. Moreover, by employing the Poincaré-Wirtinger inequality and variable substitution dexterously, the final exponential convergence criteria of the error system can be obtained with method of "Diverse treatment for same term." Finally, we provide a general numerical example and an application example in 2-D heat conduction systems to illustrate the effectiveness and practicability of the proposed measurement and control schemes.

This article attempts to design the prescribed-time time-varying deployment schemes for first-order and second-order nonlinear multiagent systems (MASs). We assume that all agents can obtain the information of their current and final relative positions with their neighbors, and the final absolute velocities (as well as their current and final relative velocities, the final absolute accelerations for the second-order MASs) through a communication network, whereas two boundary agents are able to obtain their current and final absolute positions (as well as their current and final absolute velocities for the second-order MASs). The neighbor relationship of all agents is described by a spatial variable and two static-feedback controllers are introduced, which can be expressed as a second-order space difference of the spatial variable. Then, the deployment of MASs can be transformed into the stabilization of discrete-space partial differential equation (PDE) systems. Three virtual agents are introduced to constitute the Dirchlet and Neumann boundary conditions. Several algebraic inequality criteria are derived to guarantee that the prescribed-time time-varying deployment can be achieved within a prescribed time under the Dirchlet and mixed boundary conditions. Unlike the published results, our results are derived based on the discrete-space PDE systems instead of continuous-space PDE systems, which is consistent with the discrete spatial distribution of agents. Finally, two numerical examples are given to illustrate the effectiveness of our results.

This article investigates the quasi-synchronization for fractional multiweighted coupled neural networks (FMCNNs) with discontinuous activation functions and mismatched parameters. First, under the generalized Caputo fractional-order derivative operator, a novel piecewise fractional differential inequality is established to study the convergence of fractional systems, which significantly extends some related published results. Subsequently, by exploiting the new inequality and Lyapunov stability theory, some sufficient quasi-synchronization conditions of FMCNNs are presented by aperiodic intermittent control. Meanwhile, the exponential convergence rate and synchronization error's bound are given explicitly. Finally, the validity of theoretical analysis is confirmed by numerical examples and simulations.

This article investigates the predefined-time synchronization of a group of fuzzy recurrent neural networks (FRNNs) under a leaderless communication topology. An effective control strategy is proposed based on a time-dependent exponential function as the scaling function. Sufficient criteria for guaranteeing the predefined-time synchronization of multiple FRNNs are derived under the digraph with strong connectivity and the digraph containing spanning trees, respectively. Unlike commonly used state-dependent sign function or time-dependent power function in existing works, the scaling function in this article is new and selected as the time-dependent exponential function. Moreover, the communication topology in this article is assumed to be leaderless, which is distinct from the master–slave or leader–following topologies previously investigated for predefined-time synchronization. Numerical examples are provided to illustrate the correctness of results.

This work explores the fixed-time stabilization of fuzzy uncertain systems based on conditional integral sliding mode control. The difficulties in solving such problems include the integral windup problem and the severe chattering problem when the system converges. Moreover, the design of the traditional fixed-time controller algorithm in the integral sliding mode control becomes complicated. In order to overcome these difficulties: 1) a piecewise function which can bring distant states closer to the sliding surface is considered in the integral function to prevent the integral windup problem, 2) a class of state variable controllers with odd power is designed to solve the chattering problem due to there is no sign function, 3) a fixed-time control algorithm is introduced into the integral function and the controller to guarantee that both the reachability and sliding motion phases are fixed-time stable. Furthermore, based on 1) and 3), an appropriate fixed-time conditional integral sliding mode controller is constructed to ensure the fixed-time reachability of the designed sliding mode surface. Besides, combining 1), 2) and 3), a class of chattering-free fixed-time integral sliding-mode controllers is established to obtain the fixed-time reachability of the sliding mode surface and the fixed-time stability of sliding mode dynamics. This method dramatically reduces the chattering problem and can adjust integral windup. Finally, two numerical simulations are presented to verify the theoretical results, showing the robustness of the designed overshoot factor and the chattering-free scheme for sliding mode control.

In this article, global exponential stabilization of Takagi–Sugeno (T–S) fuzzy systems with discrete time-varying delays on time scales under denial-of-service (DoS) attacks is investigated. When a DoS attack occurs, the control channel is blocked and the controller is disabled. Combining analytical method, inequality techniques, and time scale theory, stabilization criterion for the underlying systems is obtained via a fuzzy controller. Furthermore, the corresponding outcomes on continuous and discrete time domains are provided, respectively. Finally, two numerical simulations and an application of the Chua's circuit are exhibited to validate the effectiveness of the theories.

This article addresses distributed adaptive fuzzy 3-D formation tracking control of multiple autonomous underwater vehicles (AUVs) subject to marine environmental disturbances. First, by constructing a coordinate transformation, AUV model is transformed into a simple second-order systems with nonlinear dynamics. Second, fuzzy logic systems are utilized to approximate the complex nonlinear dynamics and a distributed adaptive control scheme is carried out under the assumption that all AUVs have access to the real-time information of themselves and their neighbors. Third, we assume that only sampling states of their neighbors under the event-triggered conditions are available. System states are reconstructed via the fuzzy state observers, and then the other distributed adaptive control scheme is proposed to ensure that all AUVs track the leader with the desired formation configuration under local communication. The coordinate transformation and fuzzy logic approximation method not only reduce computation but also simplify control design. Finally, a numerical simulation is carried out to demonstrate the effectiveness of the proposed control design.

This article unifies the stability criteria of asymptotic, exponential, and finite-time control within a single framework for fuzzy neural networks (FNNs) with infinite time delays. First, the boundedness and differentiability for time delays are removed. Then, Lipschitz condition for activation function is relaxed, which is allowed to have jumping discontinuous points. To stabilize FNNs, the analytical method is established by comparison principle, contradiction method and inequality techniques. Moreover, different from the traditional Lyapunov method and finite-time stability theorem, several sufficient conditions are deduced and the suppression functions are designed to guarantee asymptotic, exponential, and finite-time stabilization for FNNs by adjusting the parameters of the same controller. There is not necessary to construct the complex integral-type Lyapunov functional to deal with infinite time delays and to design power function in controller for finite-time stabilization. In addition, the designed event-triggered mechanism has the inherent advantages of saving communication resources and indirectly eliminates the chattering caused by signum function. Finally, simulations are presented to illustrate the feasibility and effectiveness of the theoretical results.

This article concentrates on the prescribed-time cooperative output regulation problem (CORP) of linear heterogeneous multiagent systems (HMASs) under directed topology. First, a novel distributed observer with prescribed-time convergence is designed to estimate the state of the exosystem. Then, two distributed prescribed-time control protocols, one based on the agent's state, the other based on the agent's output, are proposed using the observed exosystem's state. It is shown that the CORP of linear HMASs with any different orders is solved in a prescribed (any user-chosen as needed) time. Unlike the existing finite-time control strategies, the settling time is guaranteed by the designed first-order smooth control protocols, independent of the system's initial values and the control parameters. Lastly, a numerical simulation is presented to verify our theoretical results.

This article mainly explores the fixed-time control problem of fuzzy semi-Markov systems with uncertainties and unknown transition rates. Firstly, the T-S fuzzy semi-Markov system is established by using the membership relation of fuzzy logic and Markov probability property. Then, unlike the existing fixed-time control strategies, this work uses a hyperbolic sine function to replace the traditional multiple powers fixed-time control method and construct a novel fixed-time adaptive integral sliding mode control strategy, which reduces the complexity of the controller and adaptive law while optimizing the sliding mode surface and improving the fixed-time convergence performance of the system. Moreover, compared with current methods that require the assumption that the unknown function satisfies the Lipschitz condition or is bounded, the fully connected neural network is introduced to approximate the unknown nonlinear function in the system, improving the intelligence and practicality of the controller. Finally, the theoretical results are verified through numerical simulation, showing the superior performance of achieving fixed-time stability through the proposed control scheme, the gap in the study of fixed-time control using hyperbolic sine functions and fully connected neural networks is filled.

In this article, we investigate the problem of decentralized event-triggered control for large-scale systems with disturbances and input and output delays. By using the decentralized sampled-data outputs, a set of predictor-based extended state observers is first proposed to estimate the unknown states and disturbances. Thanks to the proposed predictor-based observers, the undesirable influences of both input and output delays on control performance can be effectively attenuated. With the help of disturbance estimation and attenuation technique, a set of robust sampled-data controllers is then proposed based on discrete-time event-triggering conditions, such that the disturbance rejection ability of the large-scale systems can be considerably enhanced while saving the transmission times. The stability analysis is presented to guarantee the stability of the resultant closed-loop control system, and we give the quantitative relationship of control parameters and maximum allowable intersample time interval and input and output delays. Finally, the simulation results of a numerical example are presented to verify the effectiveness of the proposed control method.

Emotion Recognition in Conversation (ERC) plays a significant part in Human-Computer Interaction (HCI) systems since it can provide empathetic services. Multimodal ERC can mitigate the drawbacks of uni-modal approaches. Recently, Graph Neural Networks (GNNs) have been widely used in a variety of fields due to their superior performance in relation modeling. In multimodal ERC, GNNs are capable of extracting both long-distance contextual information and inter-modal interactive information. Unfortunately, since existing methods such as MMGCN directly fuse multiple modalities, redundant information may be generated and diverse information may be lost. In this work, we present a directed Graph based Cross-modal Feature Complementation (GraphCFC) module that can efficiently model contextual and interactive information. GraphCFC alleviates the problem of heterogeneity gap in multimodal fusion by utilizing multiple subspace extractors and Pair-wise Cross-modal Complementary (PairCC) strategy. We extract various types of edges from the constructed graph for encoding, thus enabling GNNs to extract crucial contextual and interactive information more accurately when performing message passing. Furthermore, we design a GNN structure called GAT-MLP, which can provide a new unified network framework for multimodal learning. The experimental results on two benchmark datasets show that our GraphCFC outperforms the state-of-the-art (SOTA) approaches.

This study concentrates on the fixed-time tracking consensus and containment control of second-order heterogeneous nonlinear multiagent systems (MASs) with and without measurable velocity under directed topology. By defining a time-varying scaling function and approximating the unknown nonlinear dynamics with radial basis function neural networks (RBFNNs), a novel distributed protocol for solving the fixed-time tracking consensus and containment control problems of second-order heterogeneous nonlinear MASs with full states available is proposed based on a nonsingular sliding-mode control method constructed by designing a prescribed-time convergent sliding surface. For the scenario of immeasurable velocity, a fixed-time convergent states' observer is designed to reveal the velocity information when the unknown linearity is bounded. Subsequently, a distributed fixed-time consensus protocol based on observed velocity information is proposed for the extended results. Ultimately, the acquired results are verified by three simulation examples.

The utilization of large-scale distributed renewable energy (RE) promotes the development of the multimicrogrid (MMG), which raises the need of developing an effective energy management method to minimize economic costs and keep self energy sufficiency. The multiagent deep reinforcement learning (MADRL) has been widely used for the energy management problem because of its real-time scheduling ability. However, its training requires massive energy operation data of microgrids (MGs), while gathering these data from different MGs would threaten their privacy and data security. Therefore, this article tackles this practical yet challenging issue by proposing a federated MADRL (F-MADRL) algorithm via the physics-informed reward. In this algorithm, the federated learning (FL) mechanism is introduced to train the F-MADRL algorithm, thus ensures the privacy and the security of data. In addition, a decentralized MMG model is built, and the energy of each participated MG is managed by an agent, which aims to minimize economic costs and keep self energy sufficiency according to the physics-informed reward. At first, MGs individually execute the self-training based on local energy operation data to train their local agent models. Then, these local models are periodically uploaded to a server and their parameters are aggregated to build a global agent, which will be broadcasted to MGs and replace their local agents. In this way, the experience of each MG agent can be shared and the energy operation data are not explicitly transmitted, thus protecting the privacy and ensuring data security. Finally, experiments are conducted on Oak Ridge National Laboratory distributed energy control communication laboratory MG (ORNL-MG) test system, and the comparisons are carried out to verify the effectiveness of introducing the FL mechanism and the outperformance of our proposed F-MADRL.

This article investigates the leader-follower consensus problem of multiple Euler-Lagrange (EL) systems, where each agent suffers uncertain external disturbances, and the communication links among agents experience faults. Besides, we consider a more general case that only a portion of followers can measure partial components of leader's output and access the dynamic information of leader. The main idea of solving the consensus problem in this article is proceeded in two steps. First, we design an adaptive distributed observer to estimate the full state information of leader in real time with resilience to communication link faults. Second, based on the proposed distributed observer, we propose a proportional-integral (PI) control protocol for each agent to track the trajectory of leader, which is model-independent and robust to uncertain external disturbances. Distinct from the existing leader-follower consensus protocols of multiple EL systems, the proposed distributed observer-based PI consensus protocol in this article is model-independent, which is irrelevant to the structures or features of EL system model. Finally, we present a simulation example to show the resilience of the above adaptive distributed observer and the robustness of the distributed observer-based consensus protocol.

This article considers both the semi-Markov jumping phenomenon and spatial distribution characteristics when investigating the exponential stabilization of memristive neural networks (MNNs). The introduction of the semi-Markov jumping parameters relaxes the restriction on the sojourn time of Markovian MNNs. To increase the operability while ensuring control effect, a novel event-based spatially pointwise-piecewise switching control scheme is presented under a unified spatial division criterion, in which the pointwise and piecewise control can switch according to the preset event condition for the applicability to different control requirements. Moreover, by constructing a semi-Markov Lyapunov functional and utilizing the properties of the considered cumulative distribution function, the final exponential stabilization criterion and two related corollaries are obtained. Finally, simulation results illustrate the effectiveness and superiority of the proposed control strategy.

Global exponential stability (GES) for impulsive timescale-type nonautonomous neural networks (ITNNNs) with mixed delays is investigated in this article. Discrete time-varying and infinite distributed delays (DTVIDDs) are taken into consideration. First, an improved timescale-type Halanay inequality is proven by timescale theory. Second, several algebraic inequality criteria are demonstrated by constructing impulse-dependent functions and utilizing timescale analytical techniques. Different from the published works, the theoretical results can be applied to GES for ITNNNs and impulsive stabilization design of timescale-type nonautonomous neural networks (TNNNs) with mixed delays. The improved timescale-type Halanay inequality considers time-varying coefficients and DTVIDDs, which improves and extends some existing ones. GES criteria for ITNNNs cover the stability conditions of discrete-time nonautonomous neural networks (NNs) and continuous-time ones, and these theoretical results hold for NNs with discrete-continuous dynamics. The effectiveness of our new theoretical results is verified by two numerical examples in the end.

This article investigates the distributed formation control problem for multiple unmanned surface vessels (USVs) with model uncertainties, exogenous disturbance, and input saturation. First, a distributed finite-time sliding mode observer is developed to obtain the desired trajectory of each USV. Then, a second-order differentiable continuous fixed-time prescribed performance function is applied to reconstruct the error model based on the reference signal. In addition, the unknown part and disturbance are simultaneously handled by the composite learning control method as well as input saturation, where the acrlong NN, disturbance observer, auxiliary system, and nonsingular fast terminal sliding mode technique are integrated. Moreover, it is proved that the formation error converges to a small neighbor of the origin in a finite time. Finally, the computational simulation examples are conducted to validate the feasibility and effectiveness of the proposed method.

In this article, exponential synchronization of complex dynamical networks (CDNs) on time scales is researched. An IDET control strategy is designed to decrease the number of the event-triggered updating instants. Leveraging intermittent event detections and event-triggered sampling, and combining the analytical method with the time-scale theory, synchronization criteria are obtained for the underlying CDNs. Moreover, a parameter selection algorithm is given to acquire control parameters. In addition, two lemmas on exponential functions of time scales are proposed to prove the exclusion of Zeno behavior. Two numerical simulations and an application of formation control of spacecrafts are given to verify the validity of theoretical results.

Federated Distillation (FD) has recently attracted increasing attention for its efficiency in aggregating multiple diverse local models trained from statistically heterogeneous data of distributed clients. Existing FD methods generally treat these models equally by merely computing the average of their output soft predictions for some given input distillation sample, which does not take the diversity across all local models into account, thus leading to degraded performance of the aggregated model, especially when some local models learn little knowledge about the sample. In this paper, we propose a new perspective that treats the local data in each client as a specific domain and design a novel domain knowledge aware federated distillation method, dubbed DaFKD, that can discern the importance of each model to the distillation sample, and thus is able to optimize the ensemble of soft predictions from diverse models. Specifically, we employ a domain discriminator for each client, which is trained to identify the correlation factor between the sample and the corresponding domain. Then, to facilitate the training of the domain discriminator while saving communication costs, we propose sharing its partial parameters with the classification model. Extensive experiments on various datasets and settings show that the proposed method can improve the model accuracy by up to 6.02% compared to state-of-the-art baselines.

Partially annotated images are easy to obtain in multi-label classification. However, unknown labels in partially annotated images exacerbate the positive-negative imbalance inherent in multi-label classification, which affects supervised learning of known labels. Most current methods require sufficient image annotations, and do not focus on the imbalance of the labels in the supervised training phase. In this paper, we propose saliency regularization (SR) for a novel self-training framework. In particular, we model saliency on the class-specific maps, and strengthen the saliency of object regions corresponding to the present labels. Besides, we introduce consistency regularization to mine unlabeled information to complement unknown labels with the help of SR. It is verified to alleviate the negative dominance caused by the imbalance, and achieve state-of-the-art performance on Pascal VOC 2007, MS-COCO, VG-200, and OpenImages V3.

Compared with general optical images, hyperspectral images (HSIs) contain richer spectral information. On one hand, this provides a sufficient basis for ground object recognition. On the other hand, it results in the intermingling of spatial and spectral information. In order to make better use of the rich spatial and spectral information in HSIs, we resort to Vision Transformer (ViT). To be specific, we propose the Cross Spatial–Spectral Dense Transformer (CS2DT) for spatial-spectral feature extracting and feature fusing. For feature extraction, CS2DT employs the Adaptive Dense Encoder (ADE) module, which enables the extraction of multi-scale semantic information. During the features fusion stage, we use the Cross Spatial–Spectral Attention (CS2A) module based on the cross-attention (CA) operation to better integrate spatial and spectral features. We evaluate the classification performance of the proposed CS2DT on three well-known datasets by conducting extensive experiments. Experimental results demonstrate that CS2DT can achieve higher accuracy and higher stability when compared with the state-of-the-art (SOTA) methods. The source code will be made available at https://github.com/shouhengx/CS2DT.

Artificial neural networks are prone to suffer from catastrophic forgetting. Networks trained on something new tend to rapidly forget what was learned previously, a common phenomenon within connectionist models. In this work, we propose an effective and efficient continual learning framework using random theory, together with Bayes' rule, to equip a single model with the ability to learn streaming data. The core idea of our framework is to preserve the performance of old tasks by guiding output weights to stay in a region of low error while encountering new tasks. In contrast to the existing continual learning approaches, our main contributions concern (1) closed-formed solutions with detailed theoretical analysis; (2) training continual learners by one-pass observation of samples; (3) remarkable advantages in terms of easy implementation, efficient parameters, fast convergence, and strong task-order robustness. Comprehensive experiments under popular image classification benchmarks, FashionMNIST, CIFAR-100, and ImageNet, demonstrate that our methods predominately outperform the extensive state-of-the-art methods on training speed while maintaining superior accuracy and the number of parameters, in the class incremental learning scenario. Code is available at https://github.com/toil2sweet/CRNet.

With the increasing penetration of renewable energy and flexible loads in smart grids, a more complicated power system with high uncertainty is gradually formed, which brings about great challenges to smart grid operations. Traditional optimization methods usually require accurate mathematical models and parameters and cannot deal well with the growing complexity and uncertainty. Fortunately, the widespread popularity of advanced meters makes it possible for smart grid to collect massive data, which offers opportunities for data-driven artificial intelligence methods to address the optimal operation and control issues. Therein, deep reinforcement learning (DRL) has attracted extensive attention for its excellent performance in operation problems with high uncertainty. To this end, this article presents a comprehensive literature survey on DRL and its applications in smart grid operations. First, a detailed overview of DRL, from fundamental concepts to advanced models, is conducted in this article. Afterward, we review various DRL techniques as well as their extensions developed to cope with emerging issues in the smart grid, including optimal dispatch, operational control, electricity market, and other emerging areas. In addition, an application-oriented survey of DRL in smart grid is presented to identify difficulties for future research. Finally, essential challenges, potential solutions, and future research directions concerning the DRL applications in smart grid are also discussed.

A brain–computer interface (BCI) enables a user to communicate directly with a computer using only the central nervous system. An affective BCI (aBCI) monitors and/or regulates the emotional state of the brain, which could facilitate human cognition, communication, decision-making, and health. The last decade has witnessed rapid progress in aBCI research and applications, but there does not exist a comprehensive and up-to-date tutorial on aBCIs. This tutorial fills the gap. It introduces first the basic concepts of BCIs and then, in detail, the individual components in a closed-loop aBCI system, including signal acquisition, signal processing, feature extraction, emotion recognition, and brain stimulation. Next, it describes three representative applications of aBCIs, i.e., cognitive workload recognition, fatigue estimation, and depression diagnosis and treatment. Several challenges and opportunities in aBCI research and applications, including brain signal acquisition, emotion labeling, diversity and size of aBCI datasets, algorithm comparison, negative transfer in emotion recognition, and privacy protection and security of aBCIs, are also explained.

In this work, an operant conditioning (OC) model and memristive circuit implementation based on emotion generation and modulation is proposed, which is inspired by neural and psychological mechanisms to simulate human behavioral decisions guided by reinforcers and emotional changes during interaction with dynamically changing environments. The OC model consists of four main modules which are the cue-synapse-action module, the reward and punishment value systems, and the emotion system. Under four conditions of reward appearance or termination, and punishment appearance or termination, the memristive circuit designed according to the OC model can realize the process of exploration, acquisition, extinction, and recovery of OC (external phenomena), and also reflect the process of change of value signals and emotion signals (internal states). In addition, the phenomena of reward fatigue and punishment adaptation are considered so that value and emotion signals gradually diminish with the repetition of invariant reinforcers. The neuromorphic circuit is verified to have good robustness and device tolerance. This work enables multi-timescale OC tasks and is expected to be applied in an emotionally intelligent robot platform to achieve humanoid functions such as real-time dynamic decision-making for interaction with the environment and emotional companionship.

The geometric renormalization (GR) group of complex networks based on hidden metric space provides a powerful framework for studying the self-similarity of networks. Recent studies have shown that this framework can significantly reduce the size and complexity of the initial system. In this sense, the smaller-scale replica can be used as an alternative or guidance to the original large-scale network. In this article, we extend the GR framework to the weighted network and prove that this framework can sustain the self-similarity of synthetic weighted networks and real-world weighted networks. Furthermore, we assign the corresponding weights to all edges of reconstructed human connectomes at five different resolutions, and the results show that topological features of these networks exhibit self-similar behaviors. Remarkably, our results also suggest that the GR transform group can generate a series of low-resolution replica networks that are similar to the initial highest-resolution human connectome networks, which greatly promotes the network science, neuroscience, and physics understanding of brain mechanisms, and is of great significance to the research of brain science. Finally, a typical spin-like model is used to further verify the rationality of this framework.

The fault-tolerant consensus of nonlinear multiagent systems (MASs) is studied by using the dynamic event-triggered control and the adaptive control techniques. First, a general dynamic adaptive event-triggered mechanism (DAETM) is proposed, which promotes and improves many existing dynamic event-triggered mechanism and static event-triggered mechanism. On this basis, a new distributed dynamic adaptive event-triggered fault-tolerant controller (DDAETFTC) is designed. Then, two simple and clear criteria are derived, respectively, to ensure consensus can be reached asymptotically for nonlinear MASs with directed networks and with undirected networks under the new DDAETFTC. The obtained results also can apply to linear MASs. Furthermore, it is proven that there is no agent to show the Zeno behavior in MASs under the new DAETM. Finally, an example is given to simulate the obtained results.

In this article, the synchronization of multiple fractional-order neural networks with unbounded time-varying delays (FNNUDs) is investigated. By introducing a pinning linear control, sufficient conditions are provided for achieving the synchronization of multiple FNNUDs via an extended Halanay inequality. Moreover, a new effective adaptive control which applies to the fractional differential equations with unbounded time-varying delays is designed, under which sufficient criteria are presented to ensure the synchronization of multiple FNNUDs. The introduced control in this article is also workable in traditional integer-order neural networks. Finally, the validity of obtained results is demonstrated by a numerical example.

This article investigates the problem of periodic event-triggered output-feedback control for networked control systems in the presence of external disturbance and input and output delays. With the aid of the prediction technique, we first develop the predictor-based-extended state observer to reconstruct the system information, including the unknown state and disturbance. The periodic event-triggered output-feedback control law is then designed via the disturbance/uncertainty estimation and attenuation (DUEA) method, such that the communication times can be remarkably reduced and, at the same time, the disturbance rejection ability can be effectively enhanced. Under the predictor-based event-triggered control method, the influence of the time delays is effectively attenuated, and the effect of external disturbance is considerably attenuated due to the prediction technique and the DUEA method. By using the small-gain arguments, this article gives some sufficient stability conditions for the overall control system, and the explicit computations of sampling/updating period and time delays are presented as well. Finally, we employ a practical example and show some comparative simulation results to demonstrate the advantages of the predictor-based event-triggered control method proposed in this article.

Almost all event-triggered control (ETC) strategies were designed for discrete-time or continuous-time systems. In order to unify these existing theoretical results of ETC and develop ETC strategies for nonlinear systems, whose state variables evolve steadily at one time and change intermittently at another time, this article investigates quasisynchronization of delayed neural networks (NNs) on time scales with discontinuous activation functions via ETC approaches. First, the existence of the Filippov solutions is proved for discontinuous NNs with finite discontinuities. Second, two static event-triggered conditions and two dynamic event-triggered conditions are established to avoid continuous communication between the master-slave systems under algebraic/matrix inequality criteria. Third, under static/dynamic event-triggered conditions, a positive lower bound of event-triggered intervals is demonstrated to be greater than a positive number for each event-based controller, which shows that the Zeno behavior will not occur. Finally, two numerical simulations are carried out to show the effectiveness of the presented theoretical results in this article.

In this work, a bionic memristive circuit with the functions of emotional learning and generation is proposed, which can perform brain-like emotional learning and generation based on various types of input information. The proposed circuit is designed based on the brain emotional learning theory in the limbic system, which mainly includes three layers of design: 1) the bottom layer is the design of the basic unit modules, such as neuron and synapse; 2) the middle layer is the design of the functional modules related to emotional learning in the limbic system, such as the amygdala, thalamus, and so on; and 3) the top layer is the design of the overall circuit, which is used to realize the function of the emotional generation. A 2-D emotional space composed of valence and arousal signals is adopted. According to the above bottom-up circuit design method, the valence and arousal signals can be generated, respectively, by designing corresponding emotional learning circuits, so as to form continuous emotions. The volatile and nonvolatile memristors are mainly used to mimic the functions of the neuron and synapse at the bottom layer of the circuit to achieve the core emotional learning function of the middle layer, thereby constructing a brain-like information processing architecture to realize the function of the emotional generation in the top layer. The simulation results in PSPICE show that the proposed circuit can learn and generate emotions like humans. If the proposed circuit is applied to a humanoid robot platform through further research, the robot may have the ability of personalized emotional interaction with humans, so that it can be effectively used in emotional companionship and other aspects.

This article considers the robust consensus problem of the general linear multiagent system (MAS) subject to both heterogeneous additive stable disturbances and input saturation. Distributed low gain feedback-based dynamic output feedback control protocols are proposed, which do not need the controller interaction. Algebraic Riccati equation and unified$H_{\infty }$controller design method are employed to design the output feedback control protocols. It is established that under the assumption that each agent is asymptotically null controllable with bounded controls, semiglobal robust consensus can always be reached under the proposed controller. Furthermore, the design method specialized for leader-following consensus is addressed, under the assumption that the Laplacian matrix is diagonalizable, one can design the control protocol only with the number of follower agents. Finally, several simulations are provided to show the effectiveness of our results.

In this work, we propose a multisensory mutual associative memory networks framework and memristive circuit to mimic the ability of the biological brain to make associations of information received simultaneously. The circuit inspired by neural mechanisms of associative memory cells mainly consists of three modules: 1) the storage neurons module, which encodes external multimodal information into the firing rate of spikes; 2) the synapse module, which uses the nonvolatility memristor to achieve weight adjustment and associative learning; and 3) the retrieval neuron module, which feeds the retrieval signal output from each sensory pathway to other sensory pathways, so that achieve mutual association and retrieval between multiple modalities. Different from other one-to-one or many-to-one unidirectional associative memory work, this circuit achieves bidirectional association from multiple modalities to multiple modalities. In addition, we simulate the acquisition, extinction, recovery, transmission, and consolidation properties of associative memory. The circuit is applied to cross-modal association of image and audio recognition results, and episodic memory is simulated, where multiple images in a scene are intramodal associated. With power and area analysis, the circuit is validated as hardware-friendly. Further research to extend this work into large-scale associative memory networks, combined with visual-auditory-tactile-gustatory sensory sensors, is promising for application in intelligent robotic platforms to facilitate the development of neuromorphic systems and brain-like intelligence.

Master-slave synchronization of two delayed neural networks with adaptive controller has been studied in recent years; however, the existing delays in network models are bounded or unbounded with some derivative constraints. For more general delay without these restrictions, how to design proper adaptive controller and prove rigorously the convergence of error system is still a challenging problem. This article gives a positive answer for this problem. By means of the stability result of unbounded delayed system and some analytical techniques, we prove that the traditional centralized adaptive algorithms can achieve global asymptotical synchronization even if the network delays are unbounded without any derivative constraints. To describe the convergence speed of the synchronization error, adaptive designs depending on a flexible ω -type function are also provided to control the synchronization error, which can lead exponential synchronization, polynomial synchronization, and logarithmically synchronization. Numerical examples on delayed neural networks and chaotic Ikeda-like oscillator are presented to verify the adaptive designs, and we find that in the case of unbounded delay, the intervention of ω -type function can promote the realization of synchronization but may destroy the convergence of control gain, and this however will not happen in the case of bounded delay.

The quasisynchronization of nonidentically coupled dynamical networks (NCDNs) with proportional delay is achieved by the event-triggered mechanism (ETM). Heterogeneity and proportional delay greatly increase the difficulty on synchronization of NCDNs. As an unbounded delay in the coupling term, proportional delay is dealt with by the comparison principle, constructing parameter equations, and contradiction method. Moreover, different impulsive effects based on ETM are taken into account to reduce the burden of communication, and the quasisynchronization criteria for NCDNs are derived by the impulsive comparison principle and extended variable parameter formula. The synchronization errors and the exponential convergence rates under different impulsive effects are obtained. It is proven that the proposed ETM can avoid Zeno behavior. Finally, examples show the effectiveness of the proposed control scheme.

Seldom existing studies directly focus on the control issues of 2-D spatial partial differential equation (PDE) systems, although they have strong application backgrounds in production and life. Therefore, this article investigates the finite-time control problem of a 2-D spatial nonlinear parabolic PDE system via a Takagi–Sugeno (T–S) fuzzy boundary control scheme. First, the overall fuzzy system model is constructed using T–S fuzzy rules to approximate the original nonlinear system. Second, based on the planar distributed measurement, boundary collocated measurement, and linear measurement methods, three novel kinds of fuzzy boundary controllers are designed, respectively. Then, by employing the variable substitution and integral inequality techniques, three criteria that ensure the finite-time boundness of the considered system are obtained. Finally, simulations of main results are provided to verify the effectiveness and practicability of the proposed measurement and control schemes.

This article investigates fixed-time stabilization of fuzzy neural networks with distributed delay by designing an adaptive sliding mode controller. First, according to stability theory and related inequalities, a new fixed-time stability theorem is put forward, and the settling time is given. In order to stabilize the system, a new integral sliding mode surface is designed, and the corresponding sliding mode control strategy and adaptive sliding mode control strategy are established. Some criteria that can be obtained, and it is shown that the neuronal states of neural networks will arrive at the sliding surface in a fixed time, and then approach zero along the sliding surface. Compared with existing sliding mode control techniques, this work extends the previous related results by choosing different parameters of the controller and the sliding mode manifold to gain various protocols. Finally, two examples are provided to verify the validity of the theorems in this work.

This article is concerned with global exponential synchronization of delayed fuzzy neural networks with reaction diffusions (RDFNNs). By adopting analytic method and some inequality techniques, a global exponential synchronization criterion in terms of$p$-norm ($p\geq 2$) is obtained for the RDFNNs via adaptive intermittent control. One numerical example is provided to demonstrate the validity of the proposed outcomes. Two other examples are given to show the applications in image encryption and pseudorandom number generation, respectively.

Multimodal Emotion Recognition is challenging because of the heterogeneity gap among different modalities. Due to the powerful ability of feature abstraction, Deep Neural Networks (DNNs) have exhibited significant success in bridging the heterogeneity gap in cross-modal retrieval and generation tasks. In this work, a DNNs-based Multi-channel Weight-sharing Autoencoder with Cascade Multi-head Attention (MCWSA-CMHA) is proposed to generically address the affective heterogeneity gap in MER. Specifically, multimodal heterogeneity features are extracted by multiple independent encoders, and then a scalable heterogeneous feature fusion module (CMHA) is realized by connecting multiple multi-head attention modules in series. The core of the proposed algorithm is to reduce the heterogeneity between the output features of different encoders through the unsupervised training of MCWSA, and then to model the affective interactions between different modal features through the supervised training of CMHA. Experimental results demonstrate that the proposed MCWSA-CMHA achieves outperformance on two publicly available datasets compared with the state-of-the-art techniques. In addition, visualization experiments and approximation experiments are used to verify the effectiveness of each module in the proposed algorithm, and the experimental results show that the proposed MCWSA-CMHA can mine more emotion-related information among multimodal features compared with other fusion methods.

This article studies finite-time stabilization of delayed neural networks (DNNs) whose activation functions are discontinuous. Several sufficient conditions for guaranteeing finite-time stabilization of considered DNNs are obtained by constructing appropriate controllers with giving upper bounds of control time. Subsequently, based on the existing definition of energy consumption, the required energy to achieve stabilization is estimated. To quantify the cost of control, an evaluation index function is constructed to analyze the tradeoff between control time and consumed energy. Ultimately, acquired results are verified by simulating two numerical examples.

Recently, dynamic memristor (DM)-cellular neural networks (CNNs) have received widespread attention due to their advantage of low power consumption. The previous works showed that DM-CNNs have at most 318 equilibrium points (EPs) with$n=16$cells. Since time delay is unavoidable during the process of information transmission, the goal of this article is to research the multistability of DM-CNNs with time delay, and, meanwhile, to increase the storage capacity of DM-delay (D)CNNs. Depending on the different constitutive relations of memristors, two cases of the multistability for DM-DCNNs are discussed. After determining the constitutive relations, the number of EPs of DM-DCNNs is increased to$3^{n}$with$n$cells by means of the appropriate state-space decomposition and the Brouwer’s fixed point theorem. Furthermore, the enlarged attraction domains of EPs can be obtained, and$2^{n}$of these EPs are locally exponentially stable in two cases. Compared with standard CNNs, the dynamic behavior of DM-DCNNs shows an outstanding merit. That is, the value of voltage and current approach to zero when the system becomes stable, and the memristor provides a nonvolatile memory to store the computation results. Finally, two numerical simulations are presented to illustrate the effectiveness of the theoretical results, and the applications of associative memories are shown at the end of this article.

This article is concerned with fractional-order discontinuous complex-valued neural networks (FODCNNs). Based on a new fractional-order inequality, such system is analyzed as a compact entirety without any decomposition in the complex domain which is different from a common method in almost all literature. First, the existence of global Filippov solution is given in the complex domain on the basis of the theories of vector norm and fractional calculus. Successively, by virtue of the nonsmooth analysis and differential inclusion theory, some sufficient conditions are developed to guarantee the global dissipativity and quasi-Mittag-Leffler synchronization of FODCNNs. Furthermore, the error bounds of quasi-Mittag-Leffler synchronization are estimated without reference to the initial values. Especially, our results include some existing integer-order and fractional-order ones as special cases. Finally, numerical examples are given to show the effectiveness of the obtained theories.

This article considers the consensus problem of uncertain multiagent systems, which is addressed by neuroadaptive impulsive control schemes. The proposed control schemes indicate that the communication among agents only occurs impulsively, while the dynamics uncertainty is addressed by adaptive schemes using neural networks. Based on such approaches, two specific control schemes are designed. One is that with impulsive feedback, the control scheme uses continuous-time information, which implies that the adaptive process is continuous over time. Another is that by adopting sampled information, the update of all systems, including the feedbacks on agents, the update of neural networks, and the estimation for uncertainty, can be executed only at impulsive instants. The latter case can reduce the energy cost for communication and control, but extra assistant systems are required. The estimation and consensus prove to be achieved with errors if some conditions are fulfilled. Numerical simulations, including a practical system example, are presented.

The stability analysis of recurrent neural networks (RNNs) with multiple equilibria has received extensive interest since it is a prerequisite for successful applications of RNNs. With the increasing theoretical results on this topic, it is desirable to review the results for a systematical understanding of the state of the art. This article provides an overview of the stability results of RNNs with multiple equilibria including complete stability and multistability. First, preliminaries on the complete stability and multistability analysis of RNNs are introduced. Second, the complete stability results of RNNs are summarized. Third, the multistability results of various RNNs are reviewed in detail. Finally, future directions in these interesting topics are suggested.

This paper addresses the synchronization of multiple fractional-order recurrent neural networks (RNNs) with time-varying delays under event-triggered communications. Based on the assumption of the existence of strong connectivity or a spanning tree in the communication digraph, two sets of sufficient conditions are derived for achieving event-triggered synchronization. Moreover, an additional condition is derived to preclude Zeno behaviors. As a generalization of existing results, the criteria herein are also applicable to the event-triggered synchronization of multiple integer-order RNNs with or without delays. Two numerical examples are elaborated to illustrate the new results.

This article investigates finite-time synchronization of neural networks (NNs) with infinite discrete time-varying delays and discontinuous activations (DDNNs). By virtue of theory of differential inclusions, comparison strategies, and inequality techniques, finite-time synchronization of the underlying DDNNs can be developed via a discontinuous state feedback control law, and the synchronous settling time can be estimated. The delayed state feedback controller and finite-time stability theorem are not employed during the analysis. As a special case, finite-time synchronization of NNs with bounded delays and discontinuous activations is given. Finally, two examples are provided to illustrate the validity of the theories.

In this article, we investigate the problem of sampled-data robust output feedback control for a class of nonlinear uncertain systems with time-varying disturbance and measurement delay based on continuous-discrete observer. An augmented system that includes the nonlinear uncertain system and disturbance model is first found, and by using the delayed sampled-data output, we then propose a novel predictor-based continuous-discrete observer to estimate the unknown state and disturbance information. After that, in order to attenuate the undesirable influences of nonlinear uncertainties and disturbance, a sampled-data robust output feedback controller is developed based on disturbance/uncertainty estimation and attenuation technique. It shows that under the proposed control method, the states of overall hybrid nonlinear system can converge to a bounded region centered at the origin. The main benefit of the proposed control method is that in the presence of measurement delay, the influences of time-varying disturbance and nonlinear uncertainties can be effectively attenuated with the help of feedback domination method and prediction technique. Finally, the effectiveness of the proposed control method is demonstrated via the simulation results of a numerical example and a practical example.

This article reconsiders synchronization problem of linear complex networks with time-varying delay on time scales. For different types of time scales, aperiodically intermittent control scheme is established by using a matrix-based convex combination method, which has great potential in reducing control consumption and saving communication bandwidth. By employing a common Lyapunov function, aperiodically intermittent controllers are utilized successfully to achieve synchronization of linear delayed complex networks on special time scales onto an isolated node. Next, by constructing a special Lyapunov function with time-varying coefficients, sufficient criteria that consist of two linear matrix inequalities are demonstrated to make linear delayed complex networks on general time scales synchronized onto an isolated system with an exponential convergence rate given in advance. Due to delayed complex networks in this article defined on time scales, the proposed control schemes are applicable to continuous-time networks, their discrete-time forms, and any combination of them. Four numerical examples are offered to highlight the effectiveness and superiority of the proposed aperiodically intermittent control schemes at last.

In this article, master-slave synchronization of reaction-diffusion neural networks (RDNNs) with nondifferentiable delay is investigated via the adaptive control method. First, centralized and decentralized adaptive controllers with state coupling are designed, respectively, and a new analytical method by discussing the size of adaptive gain is proposed to prove the convergence of the adaptively controlled error system with general delay. Then, spatial coupling with adaptive gains depending on the diffusion information of the state is first proposed to achieve the master-slave synchronization of delayed RDNNs, while this coupling structure was regarded as a negative effect in most of the existing works. Finally, numerical examples are given to show the effectiveness of the proposed adaptive controllers. In comparison with the existing adaptive controllers, the proposed adaptive controllers in this article are still effective even if the network parameters are unknown and the delay is nonsmooth, and thus have a wider range of applications.

In this article, we focus on adaptive tracking control design for strict-feedback nonlinear systems with full state constraints. Unlike the barrier Lyapunov function method, all theoretical results are derived by direct method instead of introducing nonlinear state-dependent transformations, which reduces the calculated amount and the high sensitivity of control signals, and provides a new perspective to solve state-constrained problems. Then, by using the backstepping technique, the proposed adaptive controller maintains that the output closely tracks the desired trajectory, all signals in the closed-loop systems are bounded and the time-varying state constraints are not violated. Finally, two numerical simulations are provided to exhibit the effectiveness and superiority of the proposed adaptive tracking controller.

This article mainly investigates how to reduce the communication cost in multiagent system (MAS) for distributed optimization. First, a continuous-time distributed optimization model based on MAS is proposed for resource allocation (RA) with periodic communication. All agents in the system do not need to be in constant contact with their neighbors, but contact at set intervals. This will greatly reduce the communication consumption of the system. Second, to further reduce the communication cost, MAS with event-triggered communication is proposed based on periodic communication. It is proved that the system is convergent to an optimal solution of the investigated problem subject to bound and equality constraints. Finally, two examples with simulations are given to verify the performance of the proposed system.

The Halanay inequality is widely used in various time-delayed dynamical systems analyses and its vectorial form has become available recently. In this article, the integer-order vectorial Halanay-type inequality is further extended to fractional-order ones in both time-invariant and time-varying forms. It is shown that the fractional-order vectorial Halanay-type inequalities hold under the derived conditions in the form of$M$-matrices. In addition, the time-invariant inequalities are applied to analyzing the stability and synchronization of fractional-order systems with two numerical examples to substantiate the theoretical results.

In this article, we consider the disturbance rejection control problem for networked control systems using a subpredictor-based extended state observer. The networked control system is subject to large input and output delays. To handle the time delays, a subpredictor-based extended state observer is proposed to estimate the unknown state and disturbance by dividing the total delay into small pieces. The proposed observer is composed of a set of predictor-based observers connected in series, and each elementary observer is only responsible for compensating the effect of a fraction of the total delay. By using the disturbance compensation technique, we design a sampled-data composite controller to attenuate the influence of external disturbance. The proposed control method can handle the multirate case where the sensor and controller have different sampling and updating rates. It shows that under the proposed control method, the disturbance rejection ability of the control system can be effectively improved in the presence of arbitrary large input and output delays. The simulation results are given to demonstrate the effectiveness of the proposed control method.

This article focuses on the impulsive stabilization of nonautonomous timescale-type neural networks (NTNNs) with constant and unbounded time-varying delays. By choosing two discontinuous piecewise linear functions and employing the convex combination method, several algebraic criteria are demonstrated to achieve globally asymptotic stabilization of NTNNs with constant and unbounded time-varying delays. First, the asymptotic stability theorem for timescale-type systems is constructed by utilizing the timescale theory. Based on this asymptotic stability theorem, an impulsive control scheme is designed to guarantee global asymptotic stabilization of NTNNs with constant delay. Second, we propose several impulsive control schemes to achieve globally asymptotic stabilization of NTNNs with unbounded delay by virtue of the comparison strategy. Globally asymptotic stabilization criteria for NTNNs include the theoretical results of continuous-time neural networks (NNs), their discrete-time forms and NNs on continuous-discrete hybrid time scales. Especially, impulsive control for globally asymptotic stabilization of NTNNs with proportional delay is demonstrated without any variable transformation. Finally, the effectiveness of our proposed theoretical results is verified by four numerical examples.

Most synchronization criteria are scale-free on time evolution, whose main research objects are discrete-time/continuous-time systems. Unlike these theoretical results, in order to develop impulsive control schemes for discrete-time and continuous-time neural networks (NNs) in a unified framework. This article investigates impulsive control design of timescale-type NNs (TNNs) with parameter mismatches and time-varying delays (TVDs). First, several timescale impulsive differential inequalities are demonstrated by the timescale theory, which offer new inequality techniques for the investigation of timescale-type impulsive systems. Next, some criteria are proved for discrete-time NNs and TNNs by utilizing impulsive control theory, timescale inequality techniques, and the average impulsive interval method. Unlike the published works, this article gives some impulsive control schemes to ensure quasi-synchronization (QS) even if there exist TVDs in TNNs. In the end, four simulation examples are offered to demonstrate the validness of the obtained theoretical results.

DNA N6-methyladenine (6mA) is produced by the N6 position of the adenine being methylated, which occurs at the molecular level, and is involved in numerous vital biological processes in the rice genome. Given the shortcomings of biological experiments, researchers have developed many computational methods to predict 6mA sites and achieved good performance. However, the existing methods do not consider the occurrence mechanism of 6mA to extract features from the molecular structure. In this paper, a novel deep learning method is proposed by devising DNA molecular graph feature and residual block structure for 6mA sites prediction in rice, named MGF6mARice. Firstly, the DNA sequence is changed into a simplified molecular input line entry system (SMILES) format, which reflects chemical molecular structure. Secondly, for the molecular structure data, we construct the DNA molecular graph feature based on the principle of graph convolutional network. Then, the residual block is designed to extract higher level, distinguishable features from molecular graph features. Finally, the prediction module is used to obtain the result of whether it is a 6mA site. By means of 10-fold cross-validation, MGF6mARice outperforms the state-of-the-art approaches. Multiple experiments have shown that the molecular graph feature and residual block can promote the performance of MGF6mARice in 6mA prediction. To the best of our knowledge, it is the first time to derive a feature of DNA sequence by considering the chemical molecular structure. We hope that MGF6mARice will be helpful for researchers to analyze 6mA sites in rice.

Corn-plant counting is an important process for predicting corn yield and analyzing corn-plant phenotypes. In this letter, an effective corn-plant counting method is proposed, which is based on utilizing the scale-aware (SA) contextual feature and channel interdependence (CI). Given the Visual Geometry Group (VGG) Network features, the SA features are extracted by spatial pyramid pooling to derive multiscale context information. In order to utilize the channel interdependent information, the VGG features are integrated via a channel attention module. Moreover, an encoder–decoder structure is constructed to fuse the SA features and the CI-based features. Considering the sparsity of a corn plant, a hybrid loss function is adopted to train the network, by considering a density map loss function and an absolute count loss function. Experimental results demonstrate the effectiveness of the proposed method for corn-plant counting.

The functional properties of neurons in the primary visual cortex (V1) are thought to be closely related to the structural properties of this network, but the specific relationships remain unclear. Previous theoretical studies have suggested that sparse coding, an energy-efficient coding method, might underlie the orientation selectivity of V1 neurons. We thus aimed to delineate how the neurons are wired to produce this feature. We constructed a model and endowed it with a simple Hebbian learning rule to encode images of natural scenes. The excitatory neurons fired sparsely in response to images and developed strong orientation selectivity. After learning, the connectivity between excitatory neuron pairs, inhibitory neuron pairs, and excitatory-inhibitory neuron pairs depended on firing pattern and receptive field similarity between the neurons. The receptive fields (RFs) of excitatory neurons and inhibitory neurons were well predicted by the RFs of presynaptic excitatory neurons and inhibitory neurons, respectively. The excitatory neurons formed a small-world network, in which certain local connection patterns were significantly overrepresented. Bidirectionally manipulating the firing rates of inhibitory neurons caused linear transformations of the firing rates of excitatory neurons, and vice versa. These wiring properties and modulatory effects were congruent with a wide variety of data measured in V1, suggesting that the sparse coding principle might underlie both the functional and wiring properties of V1 neurons.

As an emerging in-memory element, memristor has been widely used in various neural network circuits to represent the weights and accelerate the calculation. However, the Transformer Network (TN), one of the most important models for machine vision and natural language processing in recent years, has not yet been full-circuit implemented using memristors due to the complex calculation process and data storage. In order to carry out the computation of the TN more efficiently, this work proposes a memristor-based full-circuit implementation of the TN capable of: 1) a memristor crossbar module to preserve the weights of the TN and perform the vector-matrix multiplications; 2) an analog signal memory module to store the analog signal directly in near-memory mode; 3) function circuit modules to achieve five transformations, namely Softmax, Layer Normalization, ReLU, Multiply-add and Residual; 4) a timing signal generation module to schedule operations of the circuit. The proposed TN circuit can complete all calculations directly based on the analog signal without using any analog-digital converter (ADC), digital-analog converter (DAC) and digital memory. In addition, character image recognition experiments are carried out in PSPICE to verify the functional correctness of the designed circuit. The corresponding signal retention rates of the analog memory, the performance of the whole circuit, and the non-idealities of the memristors are also analyzed. The results indicate that the circuit has advantages in terms of area overhead, energy efficiency and anti-noise.

In this article, we analyze the projective synchronization of fractional-order neural networks with mixed time delays. By introducing an extended Halanay inequality that is applicable for the case of fractional differential equations with arbitrary initial time and multiple types of delays, sufficient criteria are deduced for ensuring the projective synchronization of fractional-order neural networks with both discrete time-varying delays and distributed delays. Furthermore, sufficient criteria are presented for ensuring the projective synchronization in the Mittag-Leffler sense if there is no delay in fractional-order neural networks. The results derived herein include complete synchronization, anti-synchronization, and stabilization of fractional-order neural networks as particular cases. Moreover, the testable criteria in this article are a meaningful extension of projective synchronization of neural networks with mixed time delays from integer-order to fractional-order ones. A numerical simulation with four cases is provided to verify the validity of the obtained results.

Saturation phenomena often exist due to limited system resources, and impulsive protocols can lead to a reduction in communication cost. From these issues, this article investigates a leader-based formation control problem of multiagent systems via asynchronous impulsive protocols with saturated feedback. General linear system models with and without finite time-varying time delays under asymmetric saturated feedback control are concurrently considered. The asynchronous impulsive protocols only permit communication at impulsive instants and each agent has its own communication instants independently. Moreover, to improve system performance, an offset only containing desired formation information is introduced. Finally, because the feedbacks are saturated, admissible regions are proved to exist, which are also estimated by a mean of optimization. Numerical simulations are presented to demonstrate the validity of the proposed schemes.

The uncertainty of dynamics is often unavoidable in practical applications especially for networked control systems while energy cost reduction is a perpetual issue for engineering. With this motivation, this article considers the adaptive tracking consensus problem of uncertain second-order multiagent systems via impulsive communication. The tracking consensus is achieved by designing proper adaptive control schemes with neural networks. Two control strategies are proposed for different cases. One considers that all state information is available while another considers only partial information can be used. Estimators equipped by followers are designed, which do not need any information about neighbors during the time interval without communication. Even though the estimators and followers are under control continuously over time, the communication among all agents is only permitted at impulsive instants. In such situations, some sufficient conditions to guarantee the estimation and convergence are obtained for both cases. It is proved that by the proposed adaptive schemes for uncertain multiagent systems, errors exist both for estimation and consensus due to uncertain dynamics and adaptive schemes. Numerical simulations, including a practical example, are presented to illustrate the effectiveness of the proposed method.

These days, the synchronization of multiplex networks is an emerging and important research topic. Grounded framework and theory about synchronization and control on multiplex networks are yet to come. This article studies the intralayer synchronization on a multiplex network (i.e., a set of networks connected through interlayer edges), via the pinning impulsive control method. The topologies of different layers are independent of each other, and the individual dynamics of nodes in different layers are different as well. Supra-Laplacian matrices are adopted to represent the topological structures of multiplex networks. Two cases are considered according to impulsive sequences of multiplex networks: 1) pinning controllers are applied to all the layers simultaneously at the instants of a common impulse sequence and 2) pinning controllers are applied to each layer at the instants of distinct impulse sequences. Using the Lyapunov stability theory and the impulsive control theory, several intralayer synchronization criteria for multiplex networks are obtained, in terms of the supra-Laplacian matrix of network topology, self-dynamics of nodes, impulsive intervals, and the pinning control effect. Furthermore, the algorithms for implementing pinning schemes at every impulsive instant are proposed to support the obtained criteria. Finally, numerical examples are presented to demonstrate the effectiveness and correctness of the proposed schemes.

This article investigates the consensus control problem for second-order multiagent systems with switched dynamics, consisting of a continuous-time subsystem and a discrete-time subsystem. Under a fixed directed topology, two linear control protocols are proposed for achieving consensus. One is that two subsystems use different control inputs, where the continuous-time system uses continuous-time control, and the discrete-time system uses discrete-time control. In order to reach consensus for this kind of control protocol, some necessary and sufficient conditions are derived. The other is to use the same control algorithm for the two subsystems, which is a sampled-data control input. Similar consensus conditions are also obtained. Finally, a few simulation examples are given to verify the theoretical results.

This article proposes a Q -learning (QL)-based algorithm for global consensus of saturated discrete-time multiagent systems (DTMASs) via output feedback. According to the low-gain feedback (LGF) theory, control inputs of the saturated DTMASs can avoid the saturation by utilizing the control policies with LGF matrices, which were computed from the modified algebraic Riccati equation (MARE) by requiring the information of system dynamics in most previous works. However, in this article, we first find the lower bound on the real part of Laplacian matrices' nonzero eigenvalues of directed network topologies. Then, we define a test control input and propose a Q -function to derive a QL Bellman equation, which plays an essential part of the QL algorithm. Subsequently, different from the previous works, the output-feedback gain (OFG) matrix of this article can be obtained by limited iterations of the QL algorithm without requiring the information of agent dynamics and network topologies of the saturated DTMASs. Furthermore, the saturated DTMASs can achieve global consensus rather than the semiglobal consensus of the previous results. Finally, the effectiveness of the QL algorithm is confirmed via two simulations.

In this article, we investigate the distributed adaptive consensus problem of parabolic partial differential equation (PDE) agents by output feedback on undirected communication networks, in which two cases of no leader and leader-follower with a leader are taken into account. For the leaderless case, a novel distributed adaptive protocol, namely, the vertex-based protocol, is designed to achieve consensus by taking advantage of the relative output information of itself and its neighbors for any given undirected connected communication graph. For the case of leader-follower, a distributed continuous adaptive controller is put forward to converge the tracking error to a bounded domain by using the Lyapunov function, graph theory, and PDE theory. Furthermore, a corollary that the tracking error tends to zero by replacing the continuous controller with the discontinuous controller is given. Finally, the relevant simulation results are further demonstrated to demonstrate the theoretical results obtained.

This article investigates the global stability of bidirectional associative memory neural networks with discrete and distributed time-varying delays (DBAMNNs). By employing the comparison strategy and inequality techniques, global asymptotic stability (GAS) and global exponential stability (GES) of the underlying DBAMNNs are of concern in terms of p -norm ( p ≥ 2 ). Meanwhile, GES of the addressed DBAMNNs is also analyzed in terms of 1-norm. When distributed time delay is neglected, the GES of the corresponding bidirectional associative memory neural networks is presented as an M -matrix, which includes certain existing outcomes as special cases. Two examples are finally provided to substantiate the validity of theories.

This article investigates finite-time stabilization of competitive neural networks with discrete time-varying delays (DCNNs). By virtue of comparison strategies and inequality techniques, finite-time stabilization of the underlying DCNNs is analyzed by designing a discontinuous state feedback controller, which simplifies the controller design and proof processes of some existing results. Meanwhile, global exponential stabilization of the DCNNs is provided under a continuous state feedback controller. In addition, global exponential stability of the DCNNs is shown as an M-matrix, which contains some published outcomes as special cases. Finally, three examples are given to illuminate the validity of the theories.

Based on the fact that the traditional probability distribution entropy describing a local feature of the system cannot effectively capture the global topology variations of the network, some indicators constructed by the network adjacency matrix and Laplacian matrix come into being. Specifically, these measures are based on the eigenvalues of the scaled Laplace matrix, the eigenvalues of the network communicability matrix, and the spectral entropy based on information diffusion that has been proposed recently, respectively. In this article, we systematically study the dependence of these measures on the topological structure of the network. We prove from various aspects that spectral entropy has a better ability to identify the global topology than the traditional distribution entropy. Furthermore, the indicator based on the eigenvalues of the network communicability matrix achieves good results in some aspects while, overall, the spectral entropy is able to identify network topology variations from a global perspective.

In this article, we consider the problem of distributed adaptive leader-follower coordination of partial differential systems (i.e., reaction-diffusion neural networks, RDNNs) with directed communication topology in the case of multiple leaders. Different from the dynamical networks with ordinary differential dynamics, the design of adaptive protocols is more difficult due to the existence of spatial variables and nonlinear terms in the model. Under directed networks, a novel adaptive control protocol is proposed to solve the containment control problem of RDNNs. By constructing proper Lyapunov functional and adopting some important prior knowledge, the stability of containment for coupled RDNNs is theoretically proved. Furthermore, a corollary about the leader-follower synchronization with a leader for coupled RDNNs with directed communication topology is given. In the end, two numerical examples are provided to illustrate the obtained theoretical results.

This article is concerned with observer-based non-negative edge-consensus (OBNNEC) problems of networked discrete-time systems with or without actuator saturation. An algorithm which only uses actual outputs of neighboring edges is proposed by means of an$H_{\infty }$control method and modified algebraic Riccati equation (MARE)-based technique. The observer matrix and feedback matrix are constructed by solving the MARE and linear matrix inequality (LMI), respectively. Then, sufficient conditions for guaranteeing bounded inputs and non-negative edge states are derived. In addition, the low-gain characteristic of the MARE-based method is instrumental in guaranteeing non-negative edge states and deriving the feasible observer matrix and feedback matrix. Finally, two examples are shown to demonstrate the obtained theoretical results.

This article considers the problem of periodic event-triggered control design for dual-rate networked control systems subject to nonvanishing disturbance. The plant considered in this article is a kind of dual-rate networked control system, where the sensor samples the measurement output at a slow rate and the actuator updates the control input at a fast rate. Despite the slow-rate sampling of the sensor, a new output predictor-based observer is proposed to accurately estimate system state and disturbance in the intersample time interval, and an active anti-disturbance controller that updates at a fast rate is accordingly proposed, such that the desirable control performance and disturbance rejection performance can be achieved. At each fast-rate updating time instant, we use the prediction technique to generate a data packet, including the computed current control input and the predicted values of the control inputs for the future finite steps, and design a new periodic event-triggered mechanism to determine whether to transmit the data packet via a communication network or not. The proposed control method is easily implemented in digital platform since it has a discrete-time form. To verify the effectiveness of the proposed control method, we finally present the simulation results of a practical speed control system.

This article concerns the robust consensus problem of continuous-time linear multiagent systems (MASs) with uncertainty and discrete-time measurement information, where the output measurement information is in the data-sampled form. Distributed output-feedback protocol with or without controller interaction is proposed for each agent. Specifically, the output-feedback protocol runs in continuous time with an output error correction term mixed with the discrete-time measurement information. The concrete algorithm is given for the construction of the feedback matrices. Then, by using the delay-input approach, sufficient conditions are provided for the robust consensus of this kind of MASs interacting over networks described by the directed graphs. Finally, numerical simulations are given to illustrate the theoretical results.

In this article, we refocus on the distributed observer construction of a continuous-time linear time-invariant (LTI) system, which is called the target system, by using a network of observers to measure the output of the target system. Each observer can access only a part of the component information of the output of the target system, but the consensus-based communication among them can make it possible for each observer to estimate the full state vector of the target system asymptotically. The main objective of this article is to simplify the distributed reduced-order observer design for the LTI system on the basis of the consensus communication pattern. For observers interacting on a directed graph, we first address the problem of the distributed reduced-order observer design for the detectable target system and provide sufficient conditions involving the topology information to guarantee the existence of the distributed reduced-order observer. Then, the dependence on the topology information in the sufficient conditions will be eliminated by using the adaptive strategy and so that a completely distributed reduced-order observer can be designed for the target system. Finally, some numerical simulations are proposed to verify the theoretical results.

This article tackles the global exponential stability for a class of delayed complex-valued inertial neural networks in a discrete-time form. It is assumed that the activation function can be separated explicitly into the real part and imaginary part. Two methods are employed to deal with the stability issue. One is based on the reduced-order method. Two exponential stability criteria are obtained for the equivalent reduced-order network with the generalized matrix-measure concept. The other is directly based on the original second-order system. The main theoretical results complement each other. Some comparisons with the existing works show that the results in this article are less conservative. Two numerical examples are given to illustrate the validity of the main results.

We consider the decentralized control problem of a class of continuous-time nonlinear systems with mismatched interconnections. Initially, with the discounted cost functions being introduced to auxiliary subsystems, we have the decentralized control problem converted into a set of optimal control problems. To derive solutions to these optimal control problems, we first present the related Hamilton-Jacobi-Bellman equations (HJBEs). Then, we develop a novel critic learning method to solve these HJBEs. To implement the newly developed critic learning approach, we only use critic neural networks (NNs) and tune their weight vectors via the combination of a modified gradient descent method and concurrent learning. By using the present critic learning method, we not only remove the restriction of initial admissible control but also relax the persistence-of-excitation condition. After that, we employ Lyapunov's direct method to demonstrate that the critic NNs' weight estimation error and the states of closed-loop auxiliary systems are stable in the sense of uniform ultimate boundedness. Finally, we separately provide a nonlinear-interconnected plant and an unstable interconnected power system to validate the present critic learning approach.

This article considers the quasisynchronization of memristive neural networks (MNNs) with communication delays via event-triggered impulsive control (ETIC). In view of the limited communication and bandwidth, we adopt a novel switching event-triggered mechanism (ETM) that not only decreases the times of controller update and the amount of data sent out but also eliminates the Zeno behavior. By using an appropriate Lyapunov function, several algebraic conditions are given for quasisynchronization of MNNs with communication delays. More important, there is no restriction on the derivation of the Lyapunov function, even if it is an increasing function over a period of time. Then, we further propose a switching ETM depending on communication delays and aperiodic sampling, which is more economical and practical and can directly avoid Zeno behavior. Finally, two simulations are presented to validate the effectiveness of the proposed results.

This article solves the problems of finite-time control and settling-time estimation for fractional-order quaternion-valued fuzzy neural networks (FQFNNs) with time delays. A novel fractional differential inequality is established to estimate the settling time of the addressed system, which is more general and less conservative than the existing results. In addition, owing to the noncommutativity of multiplication of quaternions, the decomposition method is usually adopted to discuss the finite-time stabilization (FTS) of quaternion-valued neural networks, which inevitably doubles the dimensionality of the system and brings a great computational burden. To avoid the aforementioned issues, some new properties of the quaternion-valued signum function are presented for exploring the FTS of FQFNNs by the direct quaternion method without any decomposition. Then, 1-norm and 2-norm control strategies are designed to stabilize the addressed system in finite time, and several sufficient criteria are derived to ensure the FTS of FQFNNs. Finally, the validity of the obtained theoretical results and the superiority of the proposed estimation method are illustrated by numerical simulations.

The existing results of Lagrange stability for neural networks with distributed time delays are scale-free, which introduces conservativeness naturally. A class of Takagi–Sugeno fuzzy memristive neural networks (FMNNs) on time scales with discrete time-varying and infinite distributed delays is brought in this article. First, a new scale-limited Halanay inequality is demonstrated by timescale theory. Next, on the basis of inequality techniques on time scales, some new scale-limited algebraic criteria and linear matrix inequality criteria of Lagrange stability are obtained by comparison strategy and generalized Halanay inequality. All scale-limited sufficient criteria of Lagrange stability for FMNNs not only apply to continuous-time FMNNs and their discrete-time analogs, but also could deal with the arbitrary combination of them. Finally, two numerical simulations are given to verify the validity of the obtained theoretical results.

Unsupervised domain adaptation aims to learn a classification model for the target domain without any labeled samples by transferring the knowledge from the source domain with sufficient labeled samples. The source and the target domains usually share the same label space but are with different data distributions. In this paper, we consider a more difficult but insufficient-explored problem named as few-shot domain adaptation, where a classifier should generalize well to the target domain given only a small number of examples in the source domain. In such a problem, we recast the link between the source and target samples by a mixup optimal transport model. The mixup mechanism is integrated into optimal transport to perform the few-shot adaptation by learning the cross-domain alignment matrix and domain-invariant classifier simultaneously to augment the source distribution and align the two probability distributions. Moreover, spectral shrinkage regularization is deployed to improve the transferability and discriminability of the mixup optimal transport model by utilizing all singular eigenvectors. Experiments conducted on several domain adaptation tasks demonstrate the effectiveness of our proposed model dealing with the few-shot domain adaptation problem compared with state-of-the-art methods.

This article investigates the multistability and stabilization of fractional-order competitive neural networks (FOCNNs) with unbounded time-varying delays. By utilizing the monotone operator, several sufficient conditions of the coexistence of equilibrium points (EPs) are obtained for FOCNNs with concave-convex activation functions. And then, the multiple μ -stability of delayed FOCNNs is derived by the analytical method. Meanwhile, several comparisons with existing work are shown, which implies that the derived results cover the inverse-power stability and Mittag-Leffler stability as special cases. Moreover, the criteria on the stabilization of FOCNNs with uncertainty are established by designing a controller. Compared with the results of fractional-order neural networks, the obtained results in this article enrich and improve the previous results. Finally, three numerical examples are provided to show the effectiveness of the presented results.

This article investigates the stability and synchronization of nonautonomous reaction-diffusion neural networks with general time-varying delays. Compared with the existing works concerning reaction-diffusion neural networks, the main innovation of this article is that the network coefficients are time-varying, and the delays are general (which means that fewer constraints are posed on delays; for example, the commonly used conditions of differentiability and boundedness are no longer needed). By Green's formula and some analytical techniques, some easily checkable criteria on stability and synchronization for the underlying neural networks are established. These obtained results not only improve some existing ones but also contain some novel results that have not yet been reported. The effectiveness and superiorities of the established criteria are verified by three numerical examples.

This article investigates hardware implementation of hierarchical temporal memory (HTM), a brain-inspired machine learning algorithm that mimics the key functions of the neocortex and is applicable to many machine learning tasks. Spatial pooler (SP) is one of the main parts of HTM, designed to learn the spatial information and obtain the sparse distributed representations (SDRs) of input patterns. The other part is temporal memory (TM) which aims to learn the temporal information of inputs. The memristor, which is an appropriate synapse emulator for neuromorphic systems, can be used as the synapse in SP and TM circuits. In this article, a memristor-based SP (MSP) circuit structure is designed to accelerate the execution of the SP algorithm. The presented MSP has properties of modeling both the synaptic permanence and the synaptic connection state within a single synapse, and on-device and parallel learning. Simulation results of statistic metrics and classification tasks on several real-world datasets substantiate the validity of MSP.

In this article, we propose two model-free algorithms using state or output feedback for saturated discrete-time multiagent systems (SDTMASs) to attain global containment control. In most previous works, the control input can avoid saturation by utilizing the low gain feedback (LGF) method whereas requiring the knowledge of agent dynamics, and SDTMASs just can attain semi-global containment control. Distinct with the previous works, first, based on the$Q$-learning (QL) technique, this article defines a$Q$-function and deduces the corresponding QL Bellman equation, which is the most important part of the QL algorithm. Then, in order to solve the QL Bellman equation, we propose two iterative model-free algorithms using state and output feedback, and the LGF matrix can be acquired from that solution directly. Furthermore, under the state and output feedback control protocols with the feedback matrices obtained from the proposed model-free algorithms, the SDTMASs can achieve global containment control instead of semi-global containment control. Finally, we present some simulations to confirm the validity of the proposed algorithms.

In this article, we study the consensus issues of multiagent systems (MASs) without any information of the system model by using the reinforcement learning (RL) method and event-based control strategy. First, we design an adaptive event-based consensus control protocol using the local sampled state information so that the consensus errors of all agents are uniformly ultimately bounded. The validity of the above event-triggered adaptive control protocol is confirmed by excluding the Zeno behavior within finite time. Then, based on the RL approach, we present a model-free algorithm to get the feedback gain matrix, and accomplish constructing the adaptive event-triggered control strategy without the knowledge of model information. Distinct with the existing related works, this RL-based event-triggered adaptive control algorithm only relies on the local sampled state information, irrelevant to any model information or global network information. Finally, we provide some examples to demonstrate the validity of the above adaptive event-based consensus algorithm.

This article focuses on investigating the synchronization and position/force tracking performance of the multilateral teleoperation system with asymmetric time delays. First, the synchronization and tracking error signals are proposed to transform the nonlinear dynamic system into a closed-loop system governed by the Markovian jumping parameter. Because of the presence of time delays, the information regarding the system states might become unavailable. Owing to this, a feedback control based on the hidden Markov model is developed through which the information on the current state of the actual system can be accessed through a series of observations. Moreover, the forward and backward delays of the master and slave manipulators are assumed to be asymmetric and varying with time. For the stability analysis, the Lyapunov–Krasovskii technique has been adopted for which its derivatives are dealt by employing Jensens’ inequality and extended reciprocal convex matrix inequality. Finally, simulation results were provided to validate the proposed methodology guaranteeing the closed-loop system to be asymptotically stable.

This article investigates global exponential stabilization (GES) of Takagi–Sugeno (T–S) fuzzy memristive neural networks with multiple time-varying delays (DFMNNs) via intermittent control strategy. By resorting to differential inclusion theory, comparison means, and inequality techniques, some results are developed to ensure GES of the underlying DFMNNs via a fuzzy intermittent state feedback control law within the sense of Filippov. The outcome is generalized to GES of FMNNs with infinite distributed time delays. Additionally, the global exponential stability of FMNNs with discrete time-varying delays is explored in terms of 1-norm. The derived conditions herein contain certain existing ones as special cases. Finally, three examples are presented to illuminate the validness of the outcomes.

The published stability criteria for impulsive neural networks are scale-free on time line, which is only appropriate for discrete or continuous ones. The issue of global exponential stability for impulsive delayed neural networks on time scales is analyzed by employing the convex combination method in this article. Several algebraic and linear matrix inequality conditions are proved by constructing impulse-dependent Lyapunov functionals and using timescale inequality techniques. Unlike the published works, impulsive control strategies can be designed by utilizing our theoretical results to stabilize delayed neural networks on time scales if they are unstable before introducing impulses. Sufficient criteria for global exponential stability in this article are derived based on the timescale theory, and they are applicable to discrete-time impulsive neural networks, their continuous-time analogues, and neural networks whose states are discrete at one time and continuous at another time. Four numerical examples are offered to demonstrate the effectiveness and superiority of our new theoretical results in the end.

The extensively studied fractional-order dynamical networks only contain a uniform fractional derivative, which restricts the scope of fractional-order investigation. One type of Takagi–Sugeno fuzzy second-fractional-order linear networks is established in this article. First, by defining a novel Lyapunov functional, a sufficient criterion is offered to ensure global asymptotic stability of the addressed networks with time delay. Second, a fuzzy state-feedback control scheme is designed for the Takagi–Sugeno fuzzy second-fractional-order delayed linear networks to achieve global stabilization. Third, considering that only a few scholars have worked on adaptive control of fractional-order networks, an adaptive control method is also designed to realize the global stabilization of the Takagi–Sugeno fuzzy second-fractional-order linear networks, which does not contain the sign function and chattering problem can be averted. All results, given as algebraic criteria, are proved directly from the Takagi–Sugeno fuzzy second-fractional-order networks without utilizing the reduced-order approach. Three simulation examples are conducted to show the validity of the theoretical results at last.

Synchronization of a class of drive-response timescale-type nonautonomous proportional-delayed neural networks (TNPNNs) is addressed in this article. The key technique to cope with the proportional term is using comparison principle. By timescale theory, inequality technique, and comparison principle, criteria of synchronization are obtained. The method used in this article is effective to cope with TNPNNs and it is a direct approach as well by eliminating the conventional exponential transformation. The obtained results are verified with three examples.

This article concentrates on finite-time stabilization and energy consumption estimation for nonlinear systems with and without delay. By constructing an appropriate controller and utilizing inequality techniques, sufficient conditions are proposed to guarantee the finite-time stability of the delayed nonlinear system. Furthermore, the energy consumption produced in system controlling is estimated by inequality techniques. Then, we formulate similar results for the delay-free case. Finally, numerical examples are presented to demonstrate the effectiveness of our theoretical results.

In this paper, an enhanced sparse representation approach is proposed to estimate the 3D shapes of objects in 2D image sequences. In the proposed method, the unknown 3D shape is estimated via a two-stage scheme, namely the main 3D shape estimation stage and the compensatory 3D shape estimation stage. Moreover, a reweighted sparse representation model is constructed to extract the shape bases for each estimation stage. In the sparse model, a reweighted constraint is enforced to enhance the coefficient sparsity of the shape bases. Experimental results on the well-known CMU image sequences demonstrate the effectiveness and feasibility of the proposed approach.

In 1971, Dr. Leon Chua, known as the father of nonlinear circuits and cellular neural networks, postulated the existence of memristor, a portmanteau of memory resistor, in his seminal paper: Memristor-the missing circuit element published in IEEE Transactions on Circuit Theory, the predecessor of IEEE Transactions on Circuits and Systems. Thirty-seven years after he predicted its existence, in the May 1 (2008) issue of the journalNature, a team at HP Labs led by the scientist R. S. Williams proved that the memristor was real by formulating a physics-based model of a memristor and build nanoscale devices in their lab that demonstrate all of the necessary operating characteristics. Since then, the extensive interest of academic and industrial circles on neuromorphic computing based on memristor has been skyrocketed. Moreover, the unusual electrical properties of circuits and systems based on memristor can mimic the functionalities of the human brain, and can provide an in-depth understanding of key design implications of memristor-based memories, such as learning and anticipating. As a result, neuromorphic computing based on memristor is expected to bring significant breakthrough in dynamic neuromorphic memories, memristor-based resistive RAM, non-volatile memory technology, and so on.

In 1971, Dr. Leon Chua, known as the father of nonlinear circuits and cellular neural networks, postulated the existence of memristor, a portmanteau of memory resistor, in his seminal paper: “Memristor—The missing circuit element” published in IEEE Transactions on Circuit Theory, the predecessor of IEEE Transactions on Circuits and Systems—I: Regular Papers. In 2008, Hewlett-Packard researchers made nanomemristor devices for the first time, setting off an upsurge of memristor research. The emergence of nanomemristor devices is expected to realize nonvolatile RAM. Moreover, the integration, power consumption, and read–write speed of the RAM based on memristor are superior to those of traditional RAMs. The hardware network based on memristor synaptic devices is an important development direction of neuromorphic computing. It is a powerful technical candidate to break through the traditional von Neumann computing architecture in the post-Moore era, which will provide a feasible scheme about a technological breakthrough for surpassing Moore’s law.

Memristors are widely used to construct multi-scroll/wing chaotic systems with complex dynamics. However, the generation of a multi-scroll/wing attractor is typically not induced by the memristor but depends on other nonlinear functions in the system, which does not take advantage of the unique features of the memristor for chaos-based applications. To address this issue, the present paper introduces a memristor coupling (MC) method to construct a novel memristive Sprott A system (MSAS) through coupling a flux-controlled memristor with multi-piecewise linear memductance into the chaotic Sprott A system. From theoretical analysis and numerical simulations, the MSAS is shown to be able to generate any number of multi-type hidden attractors, including multi-one-scroll, multi-double-scroll and multi-double-wing hidden attractors. In addition, it has two kinds of multistabilities, that is, heterogeneous multistability and homogeneous multistability. Based on these unique properties, different numbers of coexisting heterogeneous hidden attractors and coexisting homogeneous hidden attractors are derived respectively by switching the memristor initial states. These interesting dynamical properties are comprehensively investigated using nonlinear analysis tools. Furthermore, hardware experiments are implemented to demonstrate the feasibility of the MSAS and the effectiveness of the MC method. Finally, a new pseudo-random number generator (PRNG) is proposed to explore the practical applications of the MSAS. Performance evaluation results verify the high-quality randomness of the designed PRNG.

This article discusses the synchronization problem for a class of multiple delayed neural networks (MDNNs) with a directed switching topology by using an event-triggering strategy. First, a new differential inequality with delay is shown, which is a generalization of Halanay-type inequalities. Then, the sufficient conditions of event-based synchronization (quasisynchronization) for MDNN with sequentially connected topology are obtained by using this inequality and the iterative method. Meantime, we prove that Zeno behavior can be avoided under the designed event-triggering rules. As an extension, MDNN with jointly connected topology is also discussed. Finally, a numerical example is listed to illustrate the results in theory analysis.

This article investigates the global exponential stabilization (GES) of inertial memristive neural networks with discrete and distributed time-varying delays (DIMNNs). By introducing the inertial term into memristive neural networks (MNNs), DIMNNs are formulated as the second-order differential equations with discontinuous right-hand sides. Via a variable transformation, the initial DIMNNs are rewritten as the first-order differential equations. By exploiting the theories of differential inclusion, inequality techniques, and the comparison strategy, the p th moment GES ( p ≥ 1 ) of the addressed DIMNNs is presented in terms of algebraic inequalities within the sense of Filippov, which enriches and extends some published results. In addition, the global exponential stability of MNNs is also performed in the form of an M-matrix, which contains some existing ones as special cases. Finally, two simulations are carried out to validate the correctness of the theories, and an application is developed in pseudorandom number generation.

In this article, global anti-synchronization control for a class of state-based switched inertial neural networks (SBSINNs) with time-varying delays is considered. Based on the hybrid control strategies and Lyapunov stability theory, several criteria are obtained to ensure global anti-synchronization of the underlying SBSINNs. Furthermore, we consider the global asymptotic anti-synchronization directly from the SBSINNs themselves with a nonreduced-order method. Finally, a numerical simulation is given to illustrate the effectiveness of the results.

This article addresses two kinds of formation tracking problems, namely: 1) the practical formation tracking (PFT) problem and 2) the zero-error formation tracking (ZEFT) problem for multiple Euler-Lagrange systems with input disturbances and unknown models. In these problems, the bounded input constraint, which can be possibly caused by actuator saturation and power limitations, is taken into consideration. Then, the two classes of model-independent distributed control approaches, in which the prior information (i.e., the structures and features) of the system model is not used, are proposed correspondingly. Based on the nonsmooth analysis and Lyapunov stability theory, several novel criteria for achieving PFT and ZEFT of multiple Euler-Lagrange systems are derived. Finally, numerical simulations and comparisons are presented to verify the validity and effectiveness of the proposed control approaches.

The observer design and H∞algorithm are proposed for the discrete-time fuzzy-logic systems in this article. The considered fuzzy-logic systems are subject to parameter uncertainty and unmeasurable state variables. To appropriately deal with parameter uncertainty and unmeasurable state variables, we first disintegrate the space of premise variables, then the total partitioned regions will be divided into two kinds: 1) crisp regions and 2) fuzzy regions. With the aid of partitioned regions, piecewise fuzzy H∞observers are presented. Availability of the piecewise fuzzy H∞observers gives us an accurate picture of the overall evolution of the augmented systems leading therefore to an improved situational awareness for the noncoordination of premise variables and external interference and, hence, the augmented systems achieve the performance conditions: asymptotic convergence and H∞performance. The simulation results of the proposed approach show the accuracy of the resulting state estimates.

In this article, the problems of finite-time/fixed-time synchronization have been investigated for discontinuous neural networks in the unified framework. To achieve the finite-time/fixed-time synchronization, a novel unified integral sliding-mode manifold is introduced, and corresponding unified control strategies are provided; some criteria are established for selecting suitable parameters for solving the related issue, namely, the dynamics of neural network can reach the designed sliding-mode manifold in finite/fixed time, and stay on it thereafter. Moreover, the estimations of setting time are given out. The established unified framework can bring in various protocols by choosing the different parameters of controllers and sliding-mode manifold, which extend previous related results. Finally, some numerical examples are introduced to show the effectiveness and superiority of resulting conclusions.

Existing deep neural networks (DNNs) are computationally expensive and memory intensive, which hinder their further deployment in novel nanoscale devices and applications with lower memory resources or strict latency requirements. In this paper, a novel approach to accelerate on-chip learning systems using memristive quantized neural networks (M-QNNs) is presented. A real problem of multilevel memristive synaptic weights due to device-to-device (D2D) and cycle-to-cycle (C2C) variations is considered. Different levels of Gaussian noise are added to the memristive model during each adjustment. Another method of using memristors with binary states to build M-QNNs is presented, which suffers from fewer D2D and C2C variations compared with using multilevel memristors. Furthermore, methods of solving the sneak path issues in the memristive crossbar arrays are proposed. The M-QNN approach is evaluated on two image classification datasets, that is, ten-digit number and handwritten images of mixed National Institute of Standards and Technology (MNIST). In addition, input images with different levels of zero-mean Gaussian noise are tested to verify the robustness of the proposed method. Another highlight of the proposed method is that it can significantly reduce computational time and memory during the process of image recognition.

This paper considers the self-synchronization and tracking synchronization issues for a class of nonidentically coupled neural networks model with unknown parameters and diffusion effects. Using the special structure of neural networks with global Lipschitz activation function, nonidentical terms are treated as external disturbances, which can then be compensated via robust adaptive control techniques. For the case where no common reference trajectory is given in advance, a distributed adaptive controller is proposed to drive the synchronization error to an adjustable bounded area. For the case where a reference trajectory is predesigned, two distributed adaptive controllers are proposed, respectively, to address the tracking synchronization problem with bounded and unbounded reference trajectories, different decomposition methods are given to extract the heterogeneous characteristics. To avoid the appearance of global information, such as the spectrum of the coupling matrix, corresponding adaptive designs on coupling strengths are also provided for both cases. Moreover, the upper bounds of the final synchronization errors can be gradually adjusted according to the parameters of the adaptive designs. Finally, numerical examples are given to test the effectiveness of the control algorithms.

This paper addresses the robust stability of recurrent neural networks (RNNs) with time-varying delays and input perturbation, where the time-varying delays include discrete and distributed delays. By employing the new ψ-type integral inequality, several sufficient conditions are derived for the robust stability of RNNs with discrete and distributed delays. Meanwhile, the robust boundedness of neural networks is explored by the bounded input perturbation andL1-norm constraint. Moreover, RNNs have a strong anti-jamming ability to input perturbation, and the robustness of RNNs is suitable for associative memory. Specifically, when input perturbation belongs to the specified and well-characterized space, the results cover both monostability and multistability as special cases. It is revealed that there is a relationship between the stability of neural networks and input perturbation. Compared with the existing results, these conditions proposed in this paper improve and extend the existing stability in some literature. Finally, the numerical examples are given to substantiate the effectiveness of the theoretical results.

This article mainly investigates the multiple Mittag-Leffler stability of delayed fractional-order Cohen-Grossberg neural networks with time-varying delays. By using mixed monotone operator pair, the conditions of the coexistence of multiple equilibrium points are obtained for fractional-order Cohen-Grossberg neural networks, and these conditions are eventually transformed into algebraic inequalities based on the vertex of the divided region. In particular, when the symbols of these inequalities are determined by the dominant term, several verifiable corollaries are given. And then, the sufficient conditions of the Mittag-Leffler stability are derived for fractional-order Cohen-Grossberg neural networks with time-varying delays. In addition, two numerical examples are provided to illustrate the effectiveness of the theoretical results.

This article investigates the global exponential stability and stabilization problems for a class of Takagi-Sugeno (T-S) fuzzy competitive neural networks (NNs). In the considered model, we introduce the T-S fuzzy rule to describe the parametric switching causing by complexity and the vagueness in practical environment. Besides, the effects of reaction diffusions and distributed delays, which inherently exist in circuits of NNs, are also taken into consideration. By using the Lyapunov functional theory and Green formula, several stability criteria in terms of \mathbb p-norm are established for the uncompensated fuzzy competitive NNs. Moreover, by designing a fuzzy intermittent controller, the corresponding stabilizability criteria in terms of \mathbb p-norm are derived. We also carry out some discussions and comparisons to further show the less conservativeness and wide applicability of the main theorems. Finally, several examples are presented to verify the obtained results.

As a novel nanoscale device, the memristor has elicited widespread interest in implementing compact and efficient neurocomputing systems in the hardware. In this article, fuzzy deep learning systems using fuzzy memristive modeling methods are presented. One key issue in memristive modeling is the device variation issue due to device-to-device and cycle-to-cycle variations. It is very difficult to distinguish the intermediate memristive states in a multilevel memristor. A fuzzy modeling method is therefore proposed to define the memristive states in a dynamic way. In addition, fuzzy deep learning systems are presented for fuzzy pattern recognition such as image recognition. To the authors' best knowledge, this is the first work utilizing memristive fuzzy deep learning systems to realize image recognition. The expected input and output in the memristive fuzzy deep learning systems are refined via a bidirectional fuzzy rule. The effectiveness of the proposed fuzzy methods has been verified with comprehensive methods, such as single-layer neural networks, multi-layer neural networks, convolutional neural networks, and k-nearest neighbor method. Another highlight of the proposed fuzzy deep learning system is that there is a great reduction in memory and a significant increase in the speed for image recognition tasks with the same level of testing accuracy.

Zero-shot learning has received great interest in visual recognition community. It aims to classify new unobserved classes based on the model learned from observed classes. Most zero-shot learning methods require pre-provided semantic attributes as the mid-level information to discover the intrinsic relationship between observed and unobserved categories. However, it is impractical to annotate the enriched label information of the observed objects in real-world applications, which would extremely hurt the performance of zero-shot learning with limited labeled seen data. To overcome this obstacle, we develop a Low-rank Semantics Grouping (LSG) method for zero-shot learning in a semi-supervised fashion, which attempts to jointly uncover the intrinsic relationship across visual and semantic information and recover the missing label information from seen classes. Specifically, the visual-semantic encoder is utilized as projection model, low-rank semantic grouping scheme is explored to capture the intrinsic attributes correlations and a Laplacian graph is constructed from the visual features to guide the label propagation from labeled instances to unlabeled ones. Experiments have been conducted on several standard zero-shot learning benchmarks, which demonstrate the efficiency of the proposed method by comparing with state-of-the-art methods. Our model is robust to different levels of missing label settings. Also visualized results prove that the LSG can distinguish the test unseen classes more discriminative.

This article investigates the Lagrange exponential stability and the Lyapunov exponential stability of memristive neural networks with discrete and distributed time-varying delays (DMNNs). By means of inequality techniques, theories of the M-matrix, and the comparison strategy, the Lagrange exponential stability of the underlying DMNNs is considered in the sense of Filippov, and the globally exponentially attractive set is estimated through employing the M-matrix and external input. Especially, when the external input is not concerned, the Lyapunov exponential stability of the corresponding DMNNs is developed immediately in the form of an M-matrix, which contains some published outcomes as special cases. Furthermore, by constructing an M-matrix-based differential system, the Lyapunov exponential stability of the DMNNs is studied, which is less conservative than some existing ones. Finally, three simulation examples are carried out to examine the validness of the theories.

In this article, it addresses the problem of finite-/fixed-time synchronization of delayed coupled discontinuous neural networks in the unified framework. To achieve the finite-/fixed-time synchronization and precise estimations of setting time, two novel different kinds of controllers are established, in which one is switching. Then, based on the finite-/fixed-time theorem and Lyapunov function theory, some useful criteria are obtained to select suitable controllers' parameters, which can guarantee error systems converge in the finite time/fixed time with respect to coupled neural networks. Moreover, corresponding estimations of the setting time are also provided. Finally, two numerical examples are introduced to show the effectiveness of the proposed control protocols.

This article addresses the multistability and attraction of fractional-order neural networks (FONNs) with unbounded time-varying delays. Several sufficient conditions are given to ensure the coexistence of equilibrium points (EPs) of FONNs with concave-convex activation functions. Moreover, by exploiting the analytical method and the property of the Mittag-Leffler function, it is shown that the multiple Mittag-Leffler stability of delayed FONNs is derived and the obtained criteria do not depend on differentiable time-varying delays. In particular, the criterion of the Mittag-Leffler stability can be simplified to M-matrix. In addition, the estimation of attraction basin of delayed FONNs is studied, which implies that the extension of attraction basin is independent of the magnitude of delays. Finally, three numerical examples are given to show the validity of the theoretical results.

Pinning control of a complex network aims at forcing the states of all nodes to track an external signal by controlling a small number of nodes in the network. In this paper, an algebraic graph-theoretic condition is introduced to optimize pinning control. When individual node dynamics and coupling strength of the network are given, the effectiveness of pinning scheme can be measured by the smallest eigenvalue of the grounded Laplacian matrix obtained by deleting the rows and columns corresponding to the pinned nodes from the Laplacian matrix of the network. The larger this smallest eigenvalue, the more effective the pinning scheme. Spectral properties of the smallest eigenvalue are analyzed using the network topology information, including the spectrum of the network Laplacian matrix, the minimal degree of uncontrolled nodes, the number of edges between the controlled node set and the uncontrolled node set, etc. The identified properties are shown effective for optimizing the pinning control strategy, as demonstrated by illustrative examples. Finally, for both scale-free and small-world networks, in order to maximize their corresponding smallest eigenvalues, it is better to pin the nodes with large degrees when the percentage of pinned nodes is relatively small, while it is better to pin nodes with small degrees when the percentage is relatively large. This surprising phenomenon can be explained by one of the theorems established.

This article studies the consensus of hybrid second-order multiagent systems (MASs), where the hybrid MAS is constituted by the continuous-time second-order (CTSO) and discrete-time second-order (DTSO) dynamic agents. First, a hybrid consensus algorithm is proposed, where a sample-data control is considered in the CTSO subsystems. Under the proposed hybrid consensus algorithm, the convergence of the hybrid second-order MAS matrix is analyzed. Second, a sufficient and necessary condition is proposed, which indicates that the controller parameters, such as coupling gains and the sampling interval, and eigenvalues of network topology have a significant impact on the system consensus. Finally, several simulation examples are presented to prove the validity of the results.

This paper proposes a unified framework to design sliding-mode control for stabilization of delayed memristive neural networks (DMNNs) with external disturbances. Under the presented framework, finite-time stabilization, and fixed-time stabilization of the controlled DMNNs can be, respectively, obtained by choosing different values for a specific control parameter. It is proved that the system responses can be made reaching the designed sliding-mode surface in finite and fixed time, and then stay on it. Moreover, it also illustrates that the inevitable external disturbances can be rejected by the designed sliding-mode control. Finally, the efficiency and superiority of the obtained main results are verified by comparisons with related works and numerical simulations.

This paper presents a complete solution for the hardware design of a memristor-based long short-term memory (MLSTM) network. Throughout the design process, we fully consider the external and internal structures of the long short-term memory (LSTM), both of which are efficiently implemented by memristor crossbars. In the specific design of the internal structure, the parameter sharing mechanism is used between the LSTM cells to minimize the hardware design scale. In particular, we designed a circuit that requires only one memristor crossbar for each unit in the LSTM cell. The activation function, including sigmoid and tanh (hyperbolic tangent function), involved in each unit is approximated by a piecewise function, which is designed with the corresponding hardware. To verify the effectiveness of the system we designed, we test it on IMDB and SemEval datasets. Considering the huge impact of the dimensions of the input data on the scale of the hardware design, we use word2vector instead of one-hot encoding for the input data encoding. With the parameter sharing mechanism, the transformed vectors are input in different periods, so only 65 memristive crossbars are needed in the entire system to complete the sentiment analysis of the input text. The experimental results verify the effectiveness of our proposed MLSTM system.

This paper investigates Cohen-Grossberg neural networks (CGNNs) with unbounded time-varying delays. The existence condition of multiple equilibrium points is established by the algebraic inequality method. And then, the ψ-type stability is studied by analysis method for CGNNs with unbounded time-varying delays, and the relative convergence rate can be adjusted by the selection of ψ-type functions. Moreover, by introducing the topological degree, the algebraic sum of the number of equilibria remains unchanged for any smooth function in the specified metric space. Finally, one numerical example is implemented to show the validity of the presented results.

This article deals with the asymptotic stability and synchronization of fractional-order neural networks (FONNs) with unbounded time-varying delays, in which FONNs can be the general nonlinear system. First, the existence of the equilibrium point of the nonlinear system is obtained, and the μ-type stability of fractional-order nonlinear systems is proved by the analytical method. Second, several sufficient conditions are derived to guarantee the μ-type stability of FONNs by taking easily implemented parameters. Moreover, the μ-type synchronization of drive-response systems is established by designing the linear controller. Especially, μ-type stability is the general stability, including Mittag-Leffler stability and power-stability, which is the extension of the Mittag-Leffler stability of FONNs. Finally, two numerical examples are presented to show the effectiveness of the results.

This paper investigates the quasi-synchronization of delayed memristive neural networks (MNNs) via a novel hybrid impulsive control algorithm which combines time-triggered and event-triggered impulsive control. The relationship between a predesigned non-negative auxiliary function and a given exponentially decreasing threshold function is used to describe the switching. Under this novel controller, sufficient conditions for the quasi-synchronization are derived by the impulsive differential inequality. In addition, by choosing appropriate parameters or initial conditions such that the initial value of the non-negative auxiliary function is less than that of the event-triggered function, the quasi-synchronization can be realized theoretically as long as the event-triggered impulsive intensity is less than 1. This greatly reduces the conservatism of the existing quasi-synchronization results. Furthermore, the event-triggered rules can avoid the Zeno behavior as long as the event-triggered impulsive intensity is less than 1. This hybrid mechanism can reduce the amount of impulsive control and lessen the network communication. Finally, one example is given to illustrate the validness of the obtained results.

The rate of transient faults has increased significantly as the technology scales up. The tolerance of transient faults has become an important issue in the system design. Dual modular redundancy (DMR) and triple modular redundancy (TMR) are two commonly used techniques that can achieve fault detection and masking through executing redundant tasks. As DMR and TMR have different time and cost overheads, we must carefully determine which one should be used for each task (i.e., task hardening) to achieve the optimal system design. Furthermore, for multi-core systems, the system-level design includes the allocation of cores for the tasks (i.e., task mapping) as well. This paper aims at task hardening and mapping simultaneously for independent tasks on multi-cores with heterogeneous performances, in order to minimize the maximum completion time of all tasks (i.e., makespan). We demonstrate that once task hardening is given, task mapping of independent tasks can be achieved by employing min–max-weight perfect matching with a polynomial time complexity. Besides, as there is a trade-off between cost and time performance, we propose a multi-objective memetic algorithm (MOMA)-based task hardening method to obtain a set of solutions with different numbers of cores (i.e., costs), so the designer can choose different solutions according to different requirements. The key idea of the MOMA is to incorporate problem-specific knowledge into the global search of evolutionary algorithms. Our experimental studies have demonstrated the effectiveness of the proposed method and have shown that by combining the results of MOMA and MOEA we can provide a designer with a highly accurate set of solutions within a reasonable amount of time.

This paper deals with global exponential synchronization of multiple neural networks (NNs) with time delay via a very broad class of event-triggered coupling, in which coupling matrix can be non-Laplacian. Some simple and convenient sufficient conditions are derived to guarantee global exponential synchronization of the coupling NNs under an event-triggered strategy. In particular, the effect of the common subsystem can be positive or negative on the synchronization scheme. Three examples are presented to test the results in theory analysis.

This paper addresses the problem of complete stability of delayed recurrent neural networks with a general class of piecewise linear activation functions. By applying an appropriate partition of the state space and iterating the defined bounding functions, some sufficient conditions are obtained to ensure that an n-neuron neural network is completely stable with exactly Πi=1n(2Ki-1) equilibrium points, among which Πi=1nKiequilibrium points are locally exponentially stable and the others are unstable, where Ki(i = 1, .. . ,n) are non-negative integers which depend jointly on activation functions and parameters of neural networks. The results of this paper include the existing works on the stability analysis of recurrent neural networks with piecewise linear functions as special cases and hence can be considered as the improvement and extension of the existing stability results in the literature. A numerical example is provided to illustrate the derived theoretical results.

This paper focuses on Lagrange exponential stability and finite-time stabilization of Takagi-Sugeno (T-S) fuzzy memristive neural networks with discrete and distributed time-varying delays (DFMNNs). By resorting to theories of differential inclusions and the comparison strategy, an algebraic condition is developed to confirm Lagrange exponential stability of the underlying DFMNNs in Filippov's sense, and the exponentially attractive set is estimated. When external input is not considered, global exponential stability of DFMNNs is derived directly, which includes some existing ones as special cases. Furthermore, finite-time stabilization of the addressed DFMNNs is analyzed by exploiting inequality techniques and the comparison approach via designing a nonlinear state feedback controller. The boundedness assumption of activation functions is removed herein. Finally, two simulations are presented to demonstrate the validness of the outcomes, and an application is performed in pseudorandom number generation.

In this paper, the controllability problem is addressed for a two-time-scale discrete-time system with multiple agents. First, the system is described by a singularly perturbed difference equation expressed on fast timescale. Then, to eliminate the singular perturbation parameter, by using the iterative method and approximate approach, the two-time-scale system is separated into slow and fast subsystems. Subsequently, some sufficient and/or necessary conditions of controllability for the systems are derived by using matrix theory. Moreover, under three special network topologies, the necessary criteria for controllability are proposed via graph theory. Finally, we give a simulation example to illustrate the effectiveness of the proposed theoretical results.

In this paper, a second-order consensus for multiagent systems with a directed communication topology is studied. A novel consensus strategy is first proposed, where a periodic intermittent control strategy only with casual sampled position data is used, which not only decreases the operating time and the update rates of conditioners for every individual but also responds effectively to the case of missing velocity information. A necessary and sufficient consensus condition based on the coupling gains, the sampling period, the communication width, and the spectrum of the Laplacian matrix is established to reach the consensus, and the right intervals of the sampling period are given. Furthermore, a delay-induced consensus protocol is designed, and a necessary and sufficient condition is also given, by which the sampling period and the communication width can easily be chosen to achieve the consensus. At last, some simulation examples are given to verify the correctness of the theoretical results.

The observer-based edge-consensus problem of networked continuous-time dynamical systems with edge state non-negative constraint and actuator saturation is considered in this paper. Based on line graph theory, low-gain output-feedback technique and algebraic Riccati equation (ARE)-based method, two edge-consensus algorithms are designed to achieve the observer-based edge consensus, in which the specific mathematical expressions of the two algorithms are obtained. Meanwhile, sufficient conditions are obtained to meet the bounded inputs and the non-negative edge states by combining with the ARE-based low-gain output-feedback technique and the positive system theory. Moreover, the feedback-gain and observer-gain matrices which must meet the sufficient conditions at the same time are existed and easy to obtain. Finally, two simulation cases are introduced to show the effectiveness of the theoretical results.

In this article, an event-triggered method is proposed to solve the consensus of the second-order hybrid multiagent systems (MASs), which contain discrete-time and continuous-time individuals. First, we give a selection criteria of the coupling gains, the eigenvalues of communication topology, and the event-triggered sampling interval to guarantee the hybrid consensus, which have an impact on system stability, due to the interaction and co-existence of discrete-time and continuous-time individuals. Second, the hybrid second-order consensus under the event-triggered strategy is proven, where the agents communicate with their neighbors and update their controllers only at the triggered instants. Finally, we give some simulation examples to prove the validity of the main results.

Most memristor-based Pavlov associative memory neural networks strictly require that only simultaneous food and ring appear to generate associative memory. In this article, the time delay is considered, in order to form associative memory when the food stimulus lags behind the ring stimulus for a certain period of time. In addition, the rate of learning can be changed with the length of time between the ring stimulus and food stimulus. A memristive neural network circuit that can realize Pavlov associative memory with time delay is designed and verified by the simulation results. The designed circuit consists of a synapse module, a voltage control module, and a time-delay module. The functions, such as learning, forgetting, fast learning, slow forgetting, and time-delay learning, are implemented by the circuit. The Pavlov associative memory neural network with time-delay learning provides a reference for further development of the brain-like systems.

This article investigates the global stabilization problem of Takagi-Sugeno fuzzy memristor-based neural networks with reaction-diffusion terms and distributed time-varying delays. By using the Green formula and proposing fuzzy feedback controllers, several algebraic criteria dependent on the diffusion coefficients are established to guarantee the global exponential stability of the addressed networks. Moreover, a simpler stability criterion is obtained by designing an adaptive fuzzy controller. The results derived in this article are generalized and include some existing ones as special cases. Finally, the validity of the theoretical results is verified by two examples.

A class of nonautonomous recurrent neural networks (NRNNs) with time-varying delays is considered on time scales. Bounded delays and unbounded delays have been taken into consideration, respectively. First, a new generalized Halanay inequality on time scales is constructed by time-scale theory and some analytical techniques. Based on this inequality, the stabilization of NRNNs with bounded delays is discussed on time scales. The results are also applied to the synchronization of a class of drive-response NRNNs. Furthermore, the stabilization of NRNNs with unbounded delays is investigated. Especially, the stabilization of NRNNs with proportional delays is obtained without any variable transformation. The obtained generalized Halanay inequality on time scales develops and extends some existing ones in the literature. The stabilization criteria for the NRNNs with bounded or unbounded delays cover the results of continuous-time and discrete-time NRNNs and hold the results for the systems that involved on time interval as well. Some examples are given to demonstrate the validity of the results. An application to image encryption and decryption is addressed.

In this paper, we are concerned with global stabilization for a kind of delayed memristive neural network with an inertial term. By building a new Lyapunov functional and designing a feedback controller, we obtain some new results on global stabilization of the addressed delayed memristive inertial neural networks (MINNs). An adaptive control strategy is also designed to realize the global stabilization. Compared with the reduced-order method used in the existing literature, we consider the stabilization directly from the MINNs themselves without a reduced-order method. In addition, the new results proposed here are shown as algebraic criteria, which are easy to test. At last, some simulations are given to show the validity of the derived criteria.

This paper investigates the synchronization problem of Takagi-Sugeno fuzzy memristive neural networks (FMNNs) with mixed delays, in which the bounded distributed and unbounded discrete time-varying delays are involved. Then, under the nonsmooth analysis and Lyapunov stability theory, several easily verified algebraic criteria are established to guarantee the global synchronization of FMNNs via a designed fuzzy feedback controller. Moreover, to show the superiority of the theoretical results, several discussions and comparisons with existing work are provided, indicating that derived results in this paper are general and include several existing ones as special cases. Finally, two numerical examples and two applications in psuedorandom number generation and image encryption are presented to show the validity and practicability of the theoretical results.

Fuzzy systems have achieved great success in numerous applications. However, there are still many challenges in designing an optimal fuzzy system, e.g., how to efficiently optimize its parameters, how to balance the trade-off between cooperations and competitions among the rules, how to overcome the curse of dimensionality, how to increase its generalization ability, etc. Literature has shown that by making appropriate connections between fuzzy systems and other machine learning approaches, good practices from other domains may be used to improve the fuzzy systems, and vice versa. This article gives an overview on the functional equivalence between Takagi-Sugeno-Kang fuzzy systems and four classic machine learning approaches-neural networks, mixture of experts, classification and regression trees, and stacking ensemble regression-for regression problems. We also point out some promising new research directions, inspired by the functional equivalence, that could lead to solutions to the aforementioned problems. To our knowledge, this is so far the most comprehensive overview on the connections between fuzzy systems and other popular machine learning approaches, and hopefully will stimulate more hybridization between different machine learning algorithms.

This article addresses synchronization and consensus problems in networks of linear fractional-order multi-agent systems (LFOMAS) via sampled-data control. First, under very mild assumptions, the necessary and sufficient conditions are obtained for achieving synchronization in networks of LFOMAS. Second, the results of synchronization are applied to solve some consensus problems in networks of LFOMAS. In the obtained results, the coupling matrix does not have to be a Laplacian matrix, its off-diagonal elements do not have to be nonnegative, and its row-sum can be nonzero. Finally, the validity of the theoretical results is verified by three simulation examples.

This article is devoted to the cluster synchronization issue of coupled fractional-order neural networks. By introducing the stability theory of fractional-order differential systems and the framework of Filippov regularization, some sufficient conditions are derived for ascertaining the asymptotic and finite-time cluster synchronization of coupled fractional-order neural networks, respectively. In addition, the upper bound of the settling time for finite-time cluster synchronization is estimated. Compared with the existing works, the results herein are applicable for fractional-order systems, which could be regarded as an extension of integer-order ones. A numerical example with different cases is presented to illustrate the validity of theoretical results.

In this brief, we investigate a class of second-order memristive neural networks (SMNNs) with mixed time-varying delays. Based on nonsmooth analysis, the Lyapunov stability theory, and adaptive control theory, several new results ensuring global stabilization of the SMNNs are obtained. In addition, compared with the reduced-order method used in the existing research studies, we consider the global stabilization directly from the SMNNs themselves without the reduced-order method. Finally, we give some numerical simulations to show the effectiveness of the results.

This paper investigates the φ-type stability and robust stability for a general class of stochastic reaction-diffusion neural networks (SRDNNs) with Dirichlet boundary conditions, infinite discrete time-varying delays, and infinite continuously distributed delays. By virtue of inequality techniques, properties of M-matrix, and theories of stochastic analysis, several sufficient criteria are obtained to guarantee the almost sure φ-type stability, pth moment φ-type stability, and φ-type robust stability of the underlying SRDNNs with hybrid unbounded time delays. With appropriate choices of the function φ, the φ-type stability reduces to the exponential stability, polynomial stability, and logarithmic stability. Additionally, the developed results herein include some existing ones as special cases. A numerical simulation is performed to substantiate the effectiveness and superiority of the theoretical analysis.

In this paper, a composite prescribed performance control strategy is developed for stabilizing dc/dc boost converter feeding constant power loads. First, by employing the exact feedback linearization technique, the nonlinear uncertain dc converter system is first transformed into the Brunovsky’s canonical form. Then, a nonlinear disturbance observer is utilized to evaluate the dynamic change of load power and the accuracy of output voltage regulated by feedforward compensation. Next, the prescribed performance controller is elaborately designed to ensure that the tracking error of output voltage is always within the margin of predefined error bounds. Based on the backstepping design approach, the composite nonlinear controller with prescribed performance is determined. Finally, the numerical simulation results are presented to demonstrate the tracking performance of the proposed controller.

This paper is concerned with multiple Lagrange stability under perturbation for recurrent neural networks with time-varying delays. This is different from traditional Lagrange stability, which means that multiple Lagrange stability under perturbation holds for any disturbance of initial value and any structural perturbation, within a specified and well-characterized set. In this paper, multiple Lagrange stability under perturbation with respect to a finite number of trivial solutions is established, which has good robustness. Under certain perturbation, the core of the proof of the boundedness of trajectories in mutually disjoint subregions is to employ the generalized differential inequality. The results supplement and extend some previous results, and they are applicable to robust analysis of multiple equilibria. Finally, numerical calculation and simulations are implemented to illustrate the effectiveness of the theoretical results.

For non-rigid structure from motion (NRSFM), the performance of most traditional approaches may decrease significantly when the frame number of the image sequence is relatively small. In this letter, a column space fitting (CSF) based consensus of non-rigid (CNR) reconstruction approach is proposed to deal with the 3D structure estimation problem for small-size image sequences. In the proposed method, a set of trajectory groups are first extracted by utilizing the distance weight of the pairwise points. In order to improve the estimation accuracy, an adaptive rank selection strategy is designed to choose the approximately optimal rank parameter. Corresponding to the trajectory group, the z-coordinates of the observation matrix are estimated by the CSF algorithm due to its good performance. After obtaining the outputs of the CSF-based weak estimators, the final 3D shape is derived by combining the outputs via the alternating directional method of multipliers. Experimental results on several widely used image sequences demonstrate the effectiveness and feasibility of the proposed algorithm.

In order to overcome the essential difficulties in conventional nonlinear control with iteratively adjusting multiple parameters, a novel method for designing multidirection multibutterfly chaotic attractors (MDMBCAs) without reconstructing nonlinear functions is proposed. By using a unified pulse control in a modified Lorenz system, a family of complete multibutterfly attractors can be produced, including 1-D, 2-D, and 3-D multibutterfly attractors. Theoretical analysis and numerical simulations show that arbitrary MDMBCA all can be generated by conducting the pulse-control in corresponding state variable direction (1-D), plane (2-D), or space (3-D). Meanwhile, the number of butterfly attractors can be controlled with the number of pulsed excitation. Furthermore, we design a module-based unified realization circuit and arbitrary MDMBCA can be obtained by selecting corresponding pulsed-excitation. Our theoretical analysis, MATLAB simulations and circuit experiments together show the effectiveness and universality of the proposed methodology. It should be especially pointed out that the proposed method is a universal scheme and can be applied in the arbitrary double-wing chaotic system.

Back propagation (BP) based on stochastic gradient descent is the prevailing method to train multilayer neural networks (MNNs) with hidden layers. However, the existence of the physical separation between memory arrays and arithmetic module makes it inefficient and ineffective to implement BP in conventional digital hardware. Although CMOS may alleviate some problems of the hardware implementation of MNNs, synapses based on CMOS cost too much power and areas in very large scale integrated circuits. As a novel device, memristor shows promises to overcome this shortcoming due to its ability to closely integrate processing and memory. This paper proposes a novel circuit for implementing a synapse based on a memristor and two MOSFET tansistors (p-type and n-type). Compared with a CMOS-only circuit, the proposed one reduced the area consumption by 92%-98%. In addition, we develop a fuzzy method for the adjustment of the learning rates of MNNs, which increases the learning accuracy by 2%-3% compared with a constant learning rate. Meanwhile, the fuzzy adjustment method is robust and insensitive to parameter changes due to the approximate reasoning. Furthermore, the proposed methods can be extended to memristor-based multilayer convolutional neural network for complex tasks. The novel architecture behaves in a human-liking thinking process.

Given two video frames X0and Xn+1, we aim to generate a series of intermediate frames Y1, Y2, . .., Yn, such that the resulting video consisting of frames X0, Y1- Yn, and Xn+1appears realistic to a human watcher. Such video generation has numerous important applications, including video compression, movie production, slow-motion filming, video surveillance, and forensic analysis. Yet, video generation is highly challenging due to the vast search space of possible frames. Previous methods, mostly based on video prediction and/or video interpolation, tend to generate poor-quality videos with severe motion blur. This paper proposes a novel, end-to-end approach to video generation using generative adversarial networks (GANs). In particular, our design involves two concatenated GANs, one capturing motions and the other generating frame details. The loss function is also carefully engineered to include adversarial loss, gradient difference (for motion learning), and normalized product correlation loss (for frame details). Experiments using three video datasets, namely, Google Robotic Push, KTH human actions, and UCF101, demonstrate that the proposed solution generates high-quality, realistic, and sharp videos, whereas all previous solutions output noisy and blurry results.

In this paper, the ψ -type stability and robustness of recurrent neural networks are investigated by using the differential inequality. By utilizing ψ -type functions combined with the inequality techniques, some sufficient conditions ensuring ψ -type stability and robustness are derived for linear neural networks with time-varying delays. Then, by choosing appropriate Lipschitz coefficient in subregion, some algebraic criteria of the multiple ψ -type stability and robust boundedness are established for the delayed neural networks with time-varying delays. For special cases, several criteria are also presented by selecting parameters with easy implementation. The derived results cover both ψ -type mono-stability and multiple ψ -type stability. In addition, these theoretical results contain exponential stability, polynomial stability, and μ -stability, and they also complement and extend some previous results. Finally, two numerical examples are provided to illustrate the effectiveness of the proposed criteria.

The synchronization problem of multiple/coupled reaction-diffusion neural networks with time-varying delays is investigated. Differing from the existing considerations, state delays among distinct neurons and coupling delays among different subnetworks are included in the proposed model, the assumptions posed on the arisen delays are very weak, time-varying, heterogeneous, even unbounded delays are permitted. To overcome the difficulties from this kind of delay as well as diffusion effects, a comparison-based approach is applied to this model and a series of algebraic criteria are successfully obtained to verify the global asymptotical synchronization. By specifying the existing delays, some M -matrix-based criteria are derived to justify the power-rate synchronization and exponential synchronization. In addition, new criterion on synchronization of general connected neural networks without diffusion effects is also given. Finally, two simulation examples are given to verify the effectiveness of the obtained theoretical results and provide a comparison with the existing criterion.

This paper investigates the stability and stabilization of Takagi-Sugeno (T-S) fuzzy systems with discrete and distributed time-varying delays. First, pth moment global exponential stability (p≥1) of the addressed delayed fuzzy systems is considered by virtue of the comparison approach and inequality techniques. The developed algebraic criteria include some existing outcomes as special cases. Second, global exponential stabilization of the underlying delayed fuzzy systems is performed under a fuzzy state feedback controller. Third, considering that only a few studies have been concerned with finite-time stabilization of T-S fuzzy systems, by employing the comparison strategy and a nonlinear controller, finite-time stabilization of the nominated delayed fuzzy systems is presented. The result obtained herein establishes a general theoretical framework to analyze the finite-time behavior of delayed T-S fuzzy systems. Finally, simulation examples are conducted to illustrate the validity of the results.

This paper from new perspectives discusses the global synchronization of multiple recurrent neural networks (MNNs) with time delays via impulsive coupling. A new concept (coupling strength) is introduced, it is a variable parameter and plays a key role on synchronization. The selection of coupling strength can bring more convenience to the design of the impulsive coupling controller. Four results are presented for the synchronization of MNNs with time delays by using impulsive coupling with the coupling gain and variable topology, where two results are dependent on topology and other two results are independent on topological connectivity. In our results, the effects of each NN, coupling topology, and coupling strength can be positive or negative role on synchronization. In addition, three examples are presented to test our results in the theory analysis.

This paper presents new theoretical results on the global synchronization of coupled fractional-order recurrent neural networks. Under the assumptions that the coupled fractional-order recurrent neural networks are sequentially connected in form of a single spanning tree or multiple spanning trees, two sets of sufficient conditions are derived for ascertaining the global synchronization by using the properties of Mittag-Leffler function and stochastic matrices. Compared with existing works, the results herein are applicable for fractional-order systems, which could be viewed as an extension of integer-order ones. Two numerical examples are presented to illustrate the effectiveness and characteristics of the theoretical results.

This paper is concerned with exponential stabilization for a class of Takagi-Sugeno fuzzy memristive neural networks (FMNNs) with unbounded discrete and distributed time-varying delays. Under the framework of Filippov solutions, algebraic criteria are established to guarantee exponential stabilization of the addressed FMNNs with hybrid unbounded time delays via designing a fuzzy state feedback controller by exploiting inequality techniques, calculus theorems, and theories of fuzzy sets. The obtained results in this paper enhance and generalize some existing ones. Meanwhile, a general theoretical framework is proposed to investigate the dynamical behaviors of various neural networks with mixed infinite time delays. Finally, two simulation examples are performed to illustrate the validity of the derived outcomes.

This paper considers generalized discrete-time inertial neural network (GDINN). By timescale theory, the original network is rewritten as a timescale-type inertial NN. Two different scenarios are considered. In a first scenario, several criteria guaranteeing the global exponential stability for the addressed GDINN are obtained based on the generalized matrix measure concept. In this case, Lyapunov function or functional is not necessary. In a second scenario, some inequality analytical and scaling techniques are used to achieve the global exponential stability for the considered GDINN. The obtained criteria are also applied to the global exponential synchronization of drive-response GDINNs. Several illustrative examples, including applications to the pseudorandom number generator and encrypted image transmission, are given to show the effectiveness of the theoretical results.

This paper considers the tracking synchronization problem for a class of coupled reaction-diffusion neural networks (CRDNNs) with undirected topology. For the case where the tracking trajectory has identical individual dynamic as that of the network nodes, the edge-based and vertex-based adaptive strategies on coupling strengths as well as adaptive controllers, which demand merely the local neighbor information, are proposed to synchronize the CRDNNs to the tracking trajectory. To reduce the control costs, an adaptive pinning control technique is employed. For the case where the tracking trajectory has different individual dynamic from that of the network nodes, the vertex-based adaptive strategy is proposed to drive the synchronization error to a relatively small area, which is adjustable according to the parameters of the adaptive strategy. This kind of adaptive design can enhance the robustness of the network against the external disturbance posed on the tracking trajectory. The obtained theoretical results are verified by two representative examples.

In this paper, multiple ψ -type stability of Cohen-Grossberg neural networks (CGNNs) with both time-varying discrete delays and distributed delays is investigated. By utilizing ψ -type functions combined with a new ψ -type integral inequality for treating distributed delay terms, some sufficient conditions are obtained to ensure that multiple equilibrium points are ψ -type stable for CGNNs with discrete and distributed delays, where the distributed delays include bounded and unbounded delays. These conditions of CGNNs with different output functions are less restrictive. More specifically, the algebraic criteria of the generalized model are applicable to several well-known neural network models by taking special parameters, and multiple different output functions are introduced to replace some of the same output functions, which improves the diversity of output results for the design of neural networks. In addition, the estimation of relative convergence rate of ψ -type stability is determined by the parameters of CGNNs and the selection of ψ -type functions. As a result, the existing results on multistability and monostability can be improved and extended. Finally, some numerical simulations are presented to illustrate the effectiveness of the obtained results.

This paper investigates some dynamic behaviors for a fractional-order complex-valued memristive neural networks (FCVMNNs) with delays. A new mathematical expression of the complex-value memductance (memristance) is proposed according to the feature of the complex-valued memristor-based neural networks and a new class of FCVMNNs with delays is designed. Based on the framework of Filippov solution and differential inclusion theory, the sufficient conditions are given first to guarantee the global uniform asymptotic stability for the new FCVMNNs with delays, by using fractional-order Leibniz rule and a Razumikhin-type method. In addition, a complex-value adaptive controller is designed to achieve ultimate Mittag-Leffler synchronization between two FCVMNNs with delays. Numerical simulations are given to show the effectiveness of the theoretical results.

In this paper, we propose a novel computational architecture of memristor-based echo state network (MESN) with the online least mean square (LMS) algorithm. Newman and Watts small-world network is adopted for the topological structure of MESN network with memristive neural synapses. In the MESN network, the state matrix of the reservoir layer, which is obtained by raising the dimension of input data, is utilized as an input of the LMS algorithm to train the output weight matrix on chip. After certain iterations, the resistance value of memristor is adjusted to a constant. Thus, the final weight output matrix is obtained. To verify the effectiveness of the proposed MESN network, car evaluation and short-term power load forecasting are employed with the effect evaluation of the node number and the connectivity degree of the reservoir layer. The research provides a novel way to design neuromorphic computing systems.

The existing results of passivity for neural networks mainly concentrated on first-order derivative of the states, whereas it is also significant to study the passivity with high-order derivative. In this paper, a class of memristor-based inertial neural networks (MINNs) with external inputs and outputs are concerned. First, by choosing appropriate variable transformation, the original networks are rewritten as first-order differential equations. Then a criterion of passivity for the MINNs is presented by nonsmooth analysis and linear matrix inequality (LMI) techniques, which could also result in NP-hard problem since it needs at least exponential-time to solve the passivity condition. Meanwhile, based on the obtained passivity criterion, asymptotic stability criterion is accordingly derived for MINNs. In order to avoid the NP-hard problem, we employ matrix-analysis-techniques with the property that the difference-matrix between the given matrix and the proposed matrix is seminegative definite. Then the time complexity for solving the proposed passivity criterion is reduced to constant-time with respect to the number of LMIs. Robust passivity for MINNs is also studied for bounded uncertain parameters. The MINN widens the application ranges for designing neural networks. Finally, relevant simulation examples are given to show the effectiveness of the obtained results.

In this paper, a novel nonlinear function shift method for generating multiscroll attractors is proposed, and a memristor-based control circuit is used to realize the shift controller. Three types of shift modes, namely, horizontal shift, vertical shift, and combined shift, are added in a Jerk system. The dynamic behavior is analyzed through equilibria distribution, bifurcation diagram, Lyapunov exponent spectrum, and phase portraits. Research shows that various equilibria distributions and bifurcation phenomena can be obtained by adding different shifts, thereby producing diverse attractors including periodic orbits, single-scroll, double-scroll, and multiscroll attractors. Furthermore, symmetrical and asymmetrical attractors that are unusual dynamic behaviors can also be found. The circuit construction based on CMOS technology is given, and a memristor-based control circuit is designed to implement the proposed shift method. Different multiscroll attractors can be obtained by regulating the applied control signals instead of redesigning the nonlinear circuit, which simplifies the circuit design of multiscroll system. Our theoretical analysis, numerical simulations, and PSpice simulations together demonstrate the simpleness and effectiveness of the proposed methodology.

The multivalued memory achieved with memristors is a promising approach to enhance the memory density. Effective and compact methods of reading and writing for multivalued memories can significantly improve the performance of circuits. In this paper, we present a compact and efficient scheme of reading and writing for two memristors per transistor-based multivalued memory. With the VTEAM model of the memristor, the verification of feasibility of our reading operations and writing operations for multivalued memory is achieved through HSPICE simulation.

This paper proposes a novel memristor-based multiplexer implemented by using NOT-material implication. The proposed design can be extended to arbitrary N-bit inputs and the enable-port can be added to improve its structure. Furthermore, this structure can be applied in the cascade circuit and crossbar array. A novel peripheral read circuit is introduced to overcome the problem that it is difficult to read out the operation results stored in crossbar array with large scales. The feasibility and correctness of our design is verified by the HSPICE simulation results with voltage threshold adaptive memristor model.

This paper is concerned with global asymptotic stability of delayed neural networks. Notice that a Bessel-Legendre inequality plays a key role in deriving less conservative stability criteria for delayed neural networks. However, this inequality is in the form of Legendre polynomials and the integral interval is fixed on . As a result, the application scope of the Bessel-Legendre inequality is limited. This paper aims to develop the Bessel-Legendre inequality method so that less conservative stability criteria are expected. First, by introducing a canonical orthogonal polynomial sequel, a canonical Bessel-Legendre inequality and its affine version are established, which are not explicitly in the form of Legendre polynomials. Moreover, the integral interval is shifted to a general one . Second, by introducing a proper augmented Lyapunov-Krasovskii functional, which is tailored for the canonical Bessel-Legendre inequality, some sufficient conditions on global asymptotic stability are formulated for neural networks with constant delays and neural networks with time-varying delays, respectively. These conditions are proven to have a hierarchical feature: the higher level of hierarchy, the less conservatism of the stability criterion. Finally, three numerical examples are given to illustrate the efficiency of the proposed stability criteria.

In this paper, stabilization for a class of Takagi-Sugeno (T-S) fuzzy memristive neural networks (FMNNs) with mixed time delays is investigated. By virtue of theories of differential equations with discontinuous right-hand sides, inequality techniques, and the comparison method, an algebraic criterion is derived to stabilize the addressed FMNNs with bounded discrete and distributed time delays via a designed fuzzy state feedback controller in Filippov's sense. The result can be reinforced to stabilize FMNNs with unbounded discrete time delays. Meanwhile, exponential stabilization of FMNNs with bounded discrete time delays and unbounded continuously distributed delays is also discussed. FMNNs in this study are general since fuzzy logics and hybrid time delays are all considered, and the obtained conditions enhance and extend some existing ones. Finally, four numerical simulations are carried out to substantiate the efficiency and merits of developed theoretical results.

A class of Takagi-Sugeno (T-S) fuzzy memristor-based inertial neural networks (FMINNs) is studied on time scales. The second-order derivative of the state variable in the network denotes the inertial term. At first, one timescale-type FMINNs is formulated on the basis of T-S fuzzy rules. By a variable transformation, the original network is transformed into first-order differential equations. Then, passivity criteria for the FMINNs are presented based on the characteristic function approach, linear matrix inequality techniques, and the calculus of time scales. Furthermore, two classes of control protocols, i.e., memristor- and fuzzy-related control protocols are designed to solve the passification problem for the considered FMINNs. The optimization problem of the passivity performance is also involved. Finally, simulation examples are given to show the effectiveness and validity of the obtained results, and an application is also given in pseudorandom number generation.

The existed results of Lagrange stability for neural networks (NNs) are scale-free, and hence, conservativeness appears naturally. A class of Takagi-Sugeno (T-S) fuzzy memristive NNs (FMNNs) with time-varying delays is considered on time scales. First, a class of FMNNs is formulated using characteristics of memristors and T-S fuzzy rules. Then some new scale-limited criteria of global exponential stability in Lagrange sense are obtained for FMNNs with bounded feedback functions on the basis of inequalities on time scales and inequality scaling techniques. Also, novel criteria for Lurie-type feedback functions are given, which mainly employ the constructed scale-limited generalized Halanay inequality. Moreover, by matrix-norm strategies, some matrix-norm-based scale-limited criteria are derived for bounded and Lurie-type feedback functions, respectively. It also can be seen that the matrix-norm-based criteria are in accordance with the matrix-measure-based conditions provided the time scale is specified as real set. All scale-limited criteria for Lagrange stability not only include continuous-time criteria and its discrete-time analogues, but also contain more complex cases such as the arbitrary combination of them. In the end, some numerical simulations exhibit the validity of the obtained results.

This paper is concerned with robust finite-time stabilization for a class of fractional-order neural networks (FNNs) with two types of activation functions (i.e., discontinuous and continuous activation function) under uncertainty. It is worth noting that there exist few results about FNNs with discontinuous activation functions, which is mainly because classical solutions and theories of differential equations cannot be applied in this case. Especially, there is no relevant finite-time stabilization research for such system, and this paper makes up for the gap. The existence of global solution under the framework of Filippov for such system is guaranteed by limiting discontinuous activation functions. According to set-valued analysis and Kakutani's fixed point theorem, we obtain the existence of equilibrium point. In particular, based on differential inclusion theory and fractional Lyapunov stability theory, several new sufficient conditions are given to ensure finite-time stabilization via a novel discontinuous controller, and the upper bound of the settling time for stabilization is estimated. In addition, we analyze the finite-time stabilization of FNNs with Lipschitz-continuous activation functions under uncertainty. The results of this paper improve corresponding ones of integer-order neural networks with discontinuous and continuous activation functions. Finally, three numerical examples are given to show the effectiveness of the theoretical results.

This paper is concerned with the coexistence of multiple equilibrium points and dynamical behaviors of recurrent neural networks with nonmonotonic activation functions and unbounded time-varying delays. Based on a state space partition by using the geometrical properties of the activation functions, it is revealed that an -neuron neural network can exhibit equilibrium points with . In particular, several sufficient criteria are proposed to ascertain the asymptotical stability of equilibrium points for recurrent neural networks. These theoretical results cover both monostability and multistability. Furthermore, the attraction basins of asymptotically stable equilibrium points are estimated. It is shown that the attraction basins of the stable equilibrium points can be larger than their originally partitioned subsets. Finally, the results are illustrated by using the simulation results of four examples.

This paper investigates the synchronization issue of coupled reaction-diffusion neural networks with directed topology via an adaptive approach. Due to the complexity of the network structure and the presence of space variables, it is difficult to design proper adaptive strategies on coupling weights to accomplish the synchronous goal. Under the assumptions of two kinds of special network structures, that is, directed spanning path and directed spanning tree, some novel edge-based adaptive laws, which utilized the local information of node dynamics fully are designed on the coupling weights for reaching synchronization. By constructing appropriate energy function, and utilizing some analytical techniques, several sufficient conditions are given. Finally, some simulation examples are given to verify the effectiveness of the obtained theoretical results.

Memristors have recently begun to be explored in logic operations. In this paper, a compact scheme using complementary resistive switching (CRS)-CMOS cells (CCCs) for logic operations and data storage is proposed. Several logic operations, including IMPLY, IMPLY-AND, AND, NAND, OR, and NOR, are realized with CCCs. Then the AND-OR logic and the OR-AND logic are presented to realize the programmable logic arrays built with CCCs, providing opportunities for memristor-CMOS integrated circuits.

In this paper, a dynamic aperiodic sampling approach is proposed for nonlinear model reference control systems. An exogenous dynamic model is introduced to drive the aperiodic sampling sequences. The uniformly ultimate boundedness of the reference error is proved and the proposed design is shown to be effective in reducing the sampling times. The theoretical results are validated by simulations of a nonlinear example.

This letter studies the multistability analysis of delayed recurrent neural networks with Mexican hat activation function. Some sufficient conditions are obtained to ensure that an [Formula: see text]-dimensional recurrent neural network can have [Formula: see text] equilibrium points with [Formula: see text], and [Formula: see text] of them are locally exponentially stable. Furthermore, the attraction basins of these stable equilibrium points are estimated. We show that the attraction basins of these stable equilibrium points can be larger than their originally partitioned subsets. The results of this letter improve and extend the existing stability results in the literature. Finally, a numerical example containing different cases is given to illustrate the theoretical results.

Due to the conventional look-up-table (LUT) using the static random access memory (SRAM) cell, field programmable gate arrays (FPGAs) almost reach the limitation in term of the density, speed, and configuration overhead. This paper proposes an improved memristor-based LUT (MLUT) circuit which is compatible with the mainstream LUT circuit in FPGA. Any arbitrary combined logic functions can be implemented in the MLUT through specific configurations. Then the MLUT shows superior advantages over the conventional LUT such as smaller area overhead and fewer data transmission. As a case study, a one-bit full adder is simulated to verify that the design is of practice in PSPICE. Moreover, the adder can be cascaded into multibit full adder demonstrating competitiveness against the conventional configurable logic block in FPGA technology. MLUT can be a candidate to replace the conventional SRAM-based LUT and further improves the performance of FPGAs.

This paper is concerned with synchronization for a class of reaction-diffusion neural networks with Dirichlet boundary conditions and infinite discrete time-varying delays. By utilizing theories of partial differential equations, Green's formula, inequality techniques, and the concept of comparison, algebraic criteria are presented to guarantee master-slave synchronization of the underlying reaction-diffusion neural networks via a designed controller. Additionally, sufficient conditions on exponential synchronization of reaction-diffusion neural networks with finite time-varying delays are established. The proposed criteria herein enhance and generalize some published ones. Three numerical examples are presented to substantiate the validity and merits of the obtained theoretical results.

The existed results of Lagrange stability and finite-time synchronization for memristive recurrent neural networks (MRNNs) are scale-free on time evolvement, and some restrictions appear naturally. In this paper, two novel scale-limited comparison principles are established by means of inequality techniques and induction principle on time scales. Then the results concerning Lagrange stability and global finite-time synchronization of MRNNs on time scales are obtained. Scaled-limited Lagrange stability criteria are derived, in detail, via nonsmooth analysis and theory of time scales. Moreover, novel criteria for achieving the global finite-time synchronization are acquired. In addition, the derived method can also be used to study global finite-time stabilization. The proposed results extend or improve the existed ones in the literatures. Two numerical examples are chosen to show the effectiveness of the obtained results.

This paper studies the synchronization problem of coupled delayed multistable neural networks (NNs) with directed topology. To begin with, several sufficient conditions are developed in terms of algebraic inequalities such that every subnetwork has multiple locally exponentially stable periodic orbits or equilibrium points. Then two new concepts named dynamical multisynchronization (DMS) and static multisynchronization (SMS) are introduced to describe the two novel kinds of synchronization manifolds. Using the impulsive control strategy and the Razumikhin-type technique, some sufficient conditions for both the DMS and the SMS of the controlled coupled delayed multistable NNs with fixed and switching topologies are derived, respectively. Simulation examples are presented to illustrate the effectiveness of the proposed results.

This paper addresses the issue of synchronization of switched delayed neural networks with communication delays via event-triggered control. For synchronizing coupled switched neural networks, we propose a novel event-triggered control law which could greatly reduce the number of control updates for synchronization tasks of coupled switched neural networks involving embedded microprocessors with limited on-board resources. The control signals are driven by properly defined events, which depend on the measurement errors and current-sampled states. By using a delay system method, a novel model of synchronization error system with delays is proposed with the communication delays and event-triggered control in the unified framework for coupled switched neural networks. The criteria are derived for the event-triggered synchronization analysis and control synthesis of switched neural networks via the Lyapunov-Krasovskii functional method and free weighting matrix approach. A numerical example is elaborated on to illustrate the effectiveness of the derived results.

According to conventional memristive neural network theories, neurodynamic properties are powerful tools for solving many problems in the areas of brain-like associative learning, dynamic information storage or retrieval, etc. However, as have often been noted in most fractional-order systems, system analysis approaches for integral-order systems could not be directly extended and applied to deal with fractional-order systems, and consequently, it raises difficult issues in analyzing and controlling the fractional-order memristive neural networks. By using the set-valued maps and fractional-order differential inclusions, then aided by a newly proposed fractional derivative inequality, this paper investigates the global Mittag-Leffler stabilization for a class of fractional-order memristive neural networks. Two types of control rules (i.e., state feedback stabilizing control and output feedback stabilizing control) are designed for the stabilization of fractional-order memristive neural networks, while a list of stabilization criteria is established. Finally, two numerical examples are given to show the effectiveness and characteristics of the obtained theoretical results.

In this paper, coexistence and stability of multiple equilibrium points of fractional-order recurrent neural networks are addressed. Several sufficient conditions are derived for ascertaining the existence of Πi=1n(2Ki+ 1) equilibrium points (Ki≥ 0) and the local Mittage - Leffler stability Πi=1n(Ki+ 1) equilibrium points of them by using the geometrical properties of activation functions and algebraic properties of nonsingular M-matrix. In contrast with many existing results, the derived results cover both mono-stability and multistability, and the activation functions herein could be nonmonotonic and nonlinear in any open interval. In addition, three numerical examples are elaborated to substantiate the efficacy and characteristics of the theoretical results.

This paper proposes an adaptive neural-fuzzy sliding-mode control method for uncertain nonlinear systems with actuator effectiveness faults and input saturation. The parameter dependence of the control scheme is removed from the bound of actuator faults by updating online. A neural-fuzzy model is developed to approximate the uncertain nonlinear terms and a sliding-mode online-updating controller is developed to estimate the bound of the actuator with no prior knowledge of the fault. The asymptotic stability is verified via the Lyapunov method in the presence of actuator faults and saturation. Furthermore, the adaptive neural-fuzzy control method is extended to the uncertain faulty nonlinear systems with integral sliding-mode manifold as well as other popular sliding-mode surfaces. A numerical example is presented to demonstrate the effectiveness of the derived results.

This paper investigates the fuzzy control issue for uncertain active suspension systems via dynamic sliding-mode method. The Takagi-Sugeno fuzzy approach is adopted on the background of the varying masses to describe the prescribed nonlinear system in order to achieve the design targets via the method of sector nonlinearity. This paper employs the dynamic sliding-mode scheme to control nonlinear active suspension systems. In the proposed sliding-mode control scheme, the sliding surface function is formed linearly with the system states and control inputs. Then, a fuzzy dynamic term is utilized to construct the sliding-mode feedback controller. In existing results, the sliding mode is achieved and maintained with no consideration of the system perturbations. Thus, sufficient conditions are proposed to make the sliding surface reachable with the existence of the system perturbations to make the augmented system stable. Finally, simulation results are presented to verify the effectiveness of the proposed schemes.

This paper studies the aperiodic sampled-data control for the sliding-mode control (SMC) scheme of fuzzy systems with communication-induced delays via the event-triggered method. In practice, it is impossible to update control in continuous manner; thus, an event-based control technique has become popular with the advantage that the control task is executed only if it is triggered by an event. In this paper, the event-based sliding-mode control (ESMC) is designed for each linear subsystem of the global fuzzy model first. Then, the conditions for “fuzzily” amalgamated ESMC are discussed to stabilize the global fuzzy model. This ensures that the SMC is executed only when necessary. Furthermore, the results are extended to the fuzzy systems with communication-induced delays. Finally, case studies are carried out to demonstrate the effectiveness of the derived results.

In this paper, we propose a new approach to establish a landslide displacement forecasting model based on artificial neural networks (ANNs) with random hidden weights. To quantify the uncertainty associated with the predictions, a framework for probabilistic forecasting of landslide displacement is developed. The aim of this paper is to construct prediction intervals (PIs) instead of deterministic forecasting. A lower-upper bound estimation (LUBE) method is adopted to construct ANN-based PIs, while a new single hidden layer feedforward ANN with random hidden weights for LUBE is proposed. Unlike the original implementation of LUBE, the input weights and hidden biases of the ANN are randomly chosen, and only the output weights need to be adjusted. Combining particle swarm optimization (PSO) and gravitational search algorithm (GSA), a hybrid evolutionary algorithm, PSOGSA, is utilized to optimize the output weights. Furthermore, a new ANN objective function, which combines a modified combinational coverage width-based criterion with one-norm regularization, is proposed. Two benchmark data sets and two real-world landslide data sets are presented to illustrate the capability and merit of our method. Experimental results reveal that the proposed method can construct high-quality PIs.

Recurrent neural networks, as neurodynamic systems, are a class of connectionist models that capture the dynamics of sequences via cycles in artificial neurons. Since the invention of Hopfield neural network, recurrent neural networks have attracted considerable attention, which marks the beginning of the modern age of neural network studies. Thanks to their inherent nature of parallel and distributed information processing, many computationally intensive applications can be solved by recurrent neural networks in the real-time environment.

Evaluation of uncertainties associated with landslide displacement prediction is essential for improving the reliability of landslide early warning systems. An efficient probabilistic forecasting method for the construction of prediction intervals (PIs) using bootstrap and kernel-based extreme learning machine (ELM) is proposed. To overcome the drawbacks of artificial neural networks (ANNs) in predicting mutational displacement points with time lags, this paper proposes an ANNs switched prediction scheme to construct PIs with a three-stage formulation. In the first stage, K-means clustering is applied to divide the whole training dataset into two sub-training sets: the stationary points and the mutational points. In the second stage, a weighted ELM classifier is applied to construct the switched rules. In the third stage, bootstrap- and kernel-based ELMs are applied to construct candidate PIs for each sub-training set. The final PIs are constructed by switching between these two candidate PIs. The effectiveness of the proposed ANNs switched prediction method has been validated through comprehensive tests using three real-world landslide datasets from the Three Gorges region of China.

This paper presents new theoretical results on the multistability analysis of a class of recurrent neural networks with nonmonotonic activation functions and mixed time delays. Several sufficient conditions are derived for ascertaining the existence of 3nequilibrium points and the exponential stability of 2nequilibrium points via state space partition by using the geometrical properties of activation functions and algebraic properties of nonsingular M-matrix. Compared with existing results, the conditions herein are much more computable with one order less linear matrix inequalities. Furthermore, the attraction basins of these exponentially stable equilibrium points are estimated. It is revealed that the attraction basins of the 2nequilibrium points can be larger than their originally partitioned subspaces. Three numerical examples are elaborated with typical nonmonotonic activation functions to substantiate the efficacy and characteristics of the theoretical results.

In this paper, we investigate a general class of fuzzy neurodynamic systems with piecewise constant argument of generalized type. Meanwhile, the time-varying input is under consideration. The pseudo-equilibria of this new type of neurodynamic systems is formulated and studied. Several sufficient conditions are obtained to ensure the existence and uniqueness of solutions. In addition, the global output convergence of such neurodynamic system is examined in detail. Several simulation examples are also given to verify the effectiveness of theoretical property and a potential application in analog associative memory.

This paper investigates the passivity problem for a class of uncertain stochastic fuzzy nonlinear systems with mixed delays and nonlinear noise disturbances by employing an improved free-weighting matrix approach. The fuzzy system is based on the Takagi-Sugeno model that is often used to represent the complex nonlinear systems in terms of fuzzy sets and fuzzy reasoning. To reflect more realistic dynamical behaviors of the system, the parameter uncertainties, the stochastic disturbances, and nonlinearities are considered, where the parameter uncertainties enter into all the system matrices, the stochastic disturbances are given in the form of a Brownian motion. The mixed delays comprise both discrete and distributed time-varying delays. By taking the relationship among the time delays, their lower and upper bounds into account, some less conservative linear-matrix-inequality-based delay-dependent passivity criteria are obtained without ignoring any useful terms in the derivative of Lyapunov functional. Finally, numerical examples are given to demonstrate the effectiveness and merits of the proposed methods.

This paper investigates the problem of global exponential lag synchronization of a class of switched neural networks with time-varying delays via neural activation function and applications in image encryption. The controller is dependent on the output of the system in the case of packed circuits, since it is hard to measure the inner state of the circuits. Thus, it is critical to design the controller based on the neuron activation function. Comparing the results, in this paper, with the existing ones shows that we improve and generalize the results derived in the previous literature. Several examples are also given to illustrate the effectiveness and potential applications in image encryption.

An efficient and accurate method for landslide displacement prediction is very important to reduce the casualties and property losses caused by this type of natural hazard. In recent years, many kinds of artificial neural networks (ANNs) have been widely applied to landslide displacement prediction. But we can't know which type of ANN is the best until we have calculated the prediction error. An improper choice of ANN may result in bad prediction results. In this paper, we use a neural networks combination prediction method based on the discounted MSFE (mean squared forecast error) to reduce the risk of selecting the types of ANNs. Four popular ANNs, radial basis function neural network (RBFNN), support vector regression (SVR), least squares support vector machine (LSSVM) and extreme learning machine (ELM), are selected as candidate neural networks. The performance of our model is verified through two case studies in Baishuihe landslide and Bazimen landslide. Experimental results reveal that the combining neural networks can improve the generalization abilities of ANNs.

Blood vessel segmentation is an important problem for quantitative structure analysis of retinal images, and many diseases are related to the structure changes. Manual segmentation is time consuming and computer aided segmentation is required to deal with large amount images. This paper presents a new supervised method for segmentation of blood vessels in retinal photographs. Multiple kernel learning (MKL) is introduced to deal with the problem, utilizing features from Hessian matrix based vesselness measure, response of multiscale Gabor filter, and multiple scale line strength features. The method is evaluated on the publicly available DRIVE and STARE databases. The performance of the MKL method is evaluated and experimental results show the high accuracy of the proposed method.

In this paper, a new intelligence paradigm scheme to forecast landslide based on functional networks is presented. Both methodology and learning algorithm for this kind of intelligence system paradigm using the minimax method are derived. The performance and validity of the new functional networks intelligence paradigm are demonstrated by using real-world example. The results show that the landslide prediction using functional networks is reasonable, effective and achieves a high-quality performance.