This paper proposes a fault tolerant control strategy for unmanned marine vehicles (UMVs) suffering from unknown actuator bias faults to avoid the structural complexity of conventional T-S fuzzy model. A fault observer is introduced to estimate actuator fault through online learning for real-time control input compensation. Further, to conserve network resources, an adaptive frequency event-triggered mechanism (AFETM) is applied. Unlike conventional event-triggered control methods that use fixed thresholds, AFETM dynamically adjusts the triggering condition based on historical transmission frequency and error, effectively reducing unnecessary data transmission. Moreover, instead of the traditional dual-network adaptive dynamic programming structure, a single critic network is employed to solve the Hamilton-Jacobi-Isaacs equation, significantly reducing computational burden and parameter complexity. Additionally, the stability of the UMV system is analyzed using the Lyapunov stability theorem, which demonstrates that the system states are uniformly ultimately bounded (UUB) under the proposed method. Simulation results demonstrate that the proposed AFETM reduces communication load by 99.6% and 25.4% compared with time-triggered mechanisms static and event-triggered methods, while ensuring UUB stability of the system states under actuator faults and disturbances.

As market competition intensifies and workforce scales grow, organizations increasingly seek precision in aligning role specifications with talent acquisition strategies. Employee-job alignment is a crucial aspect of strategic human resource management (HRM). Traditional methods, however, fall short in handling multi-criteria employee recommendations, adapting to dynamic preferences, and mitigating systemic biases. This study introduces a large language model (LLM)-enhanced framework that integrates dynamic hierarchical analysis with talent data retention mechanisms. A dynamic priority weight pool, based on traditional hierarchical analysis and utilizing text similarity matching, enables the LLM to efficiently access the importance matrix of skills required by the hierarchical analysis method. Simultaneously, the introduction of a bias-correction memory pool and preference aggregation pool enables the persistent storage of error correction memory and result expression bias. Leveraging the Ebbinghaus forgetting curve, a dynamic forgetting mechanism and a preference strength evaluation mechanism are developed to assess memory and preference strength, automatically eliminating redundant error correction and preference data to prevent data accumulation. Empirical evaluations show that this framework improves talent recommendation accuracy, customizes result presentation to align with institutional preferences, and reduces selection biases compared to traditional HRM systems, effectively tackling the complexities of modern workforce management.

With the popularity and development of smartphones, many mobile applications of various types have emerged. How to recommend mobile applications that match the user’s preferences and usage habits among the massive applications is a problem that needs to be solved. Traditional mobile application recommendation methods cannot dynamically track user behavior and preference changes in time and cannot timely correct the recommendation model, resulting in poor recommendation effects. The continual update of mobile applications will also invalidate the recommendation model based on traffic classification. To solve these problems, this article proposes A Deep Q-Network– (DQN) based traffic classification method for mobile application recommendation with continual learning, which embeds a DQN-based traffic classification model in the mobile terminal and sets up a reward and punishment mechanism to achieve self-supervised learning. By continuously adjusting and optimizing the model, the effectiveness of the traffic classification model is ensured, and the recommendation model is provided with accurate and reliable user behavior data support. Experiments on the ISCX and private datasets show that the proposed method performs better and can effectively guarantee the accuracy of the classification model.

Many commonly used video captioning datasets contain multiple caption annotations per video. When training with cross-entropy loss, the model encounters ambiguity because the same input is mapped to different targets, leading to confusion. To address this issue, we propose the Multi-label Semantic Learning and Relay (MSLR) framework, which transforms the one-to-many relation between videos and their descriptions into a one-to-one mapping. Specifically, MSLR introduces two modules after the decoder. The Multi-label Multi-granularity Learning (MML) module integrates sentence-level granularity from multiple descriptions and employs attention mechanisms to extract word-level granularity weights, thereby capturing complementary semantic information from diverse perspectives. The Multi-label Semantic Relay (MSR) module subsequently leverages a parameter-sharing mechanism to feed complementary semantics back into the decoder, thus preventing the generation of overly generic descriptions during inference. Compared to state-of-the-art lightweight methods, MSLR achieves highly competitive results. The code is available at https://github.com/hyk0320/MSLR.

Large-scale vision foundation models such as DINOv2 boast impressive performances by leveraging massive architectures and training datasets. The expense of large-scale pre-training puts such research out of reach for many, hence limiting scientific advancements. We thus propose a novel pretraining strategy for DINOv2 that simultaneously accelerates convergence–and strengthens robustness to common corruptions as a by-product. Our approach involves a frequency filtering curriculum–low-frequency being seen first–and the Gaussian noise patching augmentation. Applied to a ViT-B/16 backbone trained on ImageNet-1K, while pre-training time is reduced by 1.6×–from 16.64 to 10.32 NVIDIA L40S days–and FLOPs by 2.25×, our method still achieves matching robustness in corruption benchmarks (ImageNet-C) and maintains competitive linear probing performance compared with the DINOv2 baseline. This dual benefit of efficiency and robustness makes large-scale self-supervised foundation modeling more attainable, while opening the door to novel exploration around data curriculum and augmentation as a means to improve self-supervised learning models robustness.

Social media platforms such as Weibo and TikTok have become more influential than traditional media. Sentiment in social media comments reflects users’ attitudes and impacts society, making sentiment analysis (SA) crucial. AI driven models, especially deep-learning models, have achieved excellent results in SA tasks. However, most existing models are not interpretable enough. First, deep learning models have numerous parameters, and their transparency is insufficient. People cannot easily understand how the models extract features from input data and make sentiment judgments. Second, most models lack intuitive explanations. They cannot clearly indicate which words or phrases are key for emotion prediction. Moreover, extracting sentiment factors from comments is challenging because a comment often contains multiple sentiment characteristics. To address these issues, we propose a dual-branch combination network (DCN) for SA of social media short comments, achieving both word-level and sentence-level interpretability. The network includes a key word feature extraction network (KWFEN) and a key word order feature extraction network (KWOFEN). KWFEN uses popular emotional words and SHAP for word-level interpretability. KWOFEN employs position embedding and an attention layer to visualize attention weights for sentence-level interpretability. We validated our method on the public dataset weibo2018 and TSATC. The results show that our method effectively extracts positive and negative sentiment factors, establishing a clear mapping between model inputs and outputs, demonstrating good interpretability performance.

Deformable image registration (DIR) is an active research topic in biomedical imaging. There is a growing interest in developing DIR methods based on deep learning (DL). A traditional DL approach to DIR is based on training a convolutional neural network (CNN) to estimate the registration field between two input images. While conceptually simple, this approach comes with a limitation that it exclusively relies on a pre-trained CNN without explicitly enforcing fidelity between the registered image and the reference. We present plug-and-play image registration network (PIRATE) as a new DIR method that addresses this issue by integrating an explicit data-fidelity penalty and a CNN prior. PIRATE pre-trains a CNN denoiser on the registration field and "plugs" it into an iterative method as a regularizer. We additionally present PIRATE+ that fine-tunes the CNN prior in PIRATE using deep equilibrium models (DEQ). PIRATE+ interprets the fixed-point iteration of PIRATE as a network with effectively infinite layers and then trains the resulting network end-to-end, enabling it to learn more task-specific information and boosting its performance. Our numerical results on OASIS and CANDI datasets show that our methods achieve state-of-the-art performance on DIR.

We present Jade, a differentiable physics engine for articulated rigid bodies. Jade models contacts as the Linear Complementarity Problem (LCP). Compared to existing differentiable simulations, Jade offers features including intersection-free collision simulation and stable LCP solutions for multiple frictional contacts. We use continuous collision detection to detect the time of impact and adopt the backtracking strategy to prevent intersection between bodies with complex geometry shapes. We derive the gradient calculation to ensure the whole simulation process is differentiable under the backtracking mechanism. We modify the popular Dantzig’s algorithm to get valid solutions under multiple frictional contacts. We conduct extensive experiments to demonstrate the effectiveness of our differentiable physics simulation over a variety of contact-rich tasks. Supplemental materials and videos are available on our project webpage at https://sites.google.com/view/diffsim

Authentication is crucial for network system security, relying on methods such as passwords, ID cards, biometrics, and behavioral characteristics. The conventional centralized authentication may lead to potential performance bottlenecks and privacy risks such as key exposure, single point of failure. Decentralized authentication systems using cryptographic techniques aim to address these issues but often tradeoff between flexibility and communication efficiency. In this paper we propose a new cryptographic concept called designated private set-based trapdoor authentication (DPSBTA) for flexible and efficient trust management in decentralized systems. DPSBTA eliminates the need for a trusted authority, with users’ access privileges defined by their private sets. During the authentication process, each server can designate an element set and only if a user holds adequate elements which are contained in the designated set can he obtains a credential from the server. The key features of DPSBTA include: decentralized trapdoor authentication management, without a trusted authority, conducted in a double threshold manner; privacy preservation, as servers do not know users’ element holdings or credential generation; round-optimal communication, with only two rounds of interaction between users and servers. We present the generic construction, security models, and concrete algorithms with correctness proof. The theoretical proof and the performance evaluations demonstrate the tangible security and high efficacy of the proposed DPSBTA.

Implicit neural representations (INR) have been recently proposed as deep learning (DL) based solutions for image compression. An image can be compressed by training an INR model with fewer weights than the number of image pixels to map the coordinates of the image to corresponding pixel values. While traditional training approaches for INRs are based on enforcing pixel-wise image consistency, we propose to further improve image quality by using a new structural regularizer. We present structural regularization for INR compression (SINCO) as a novel INR method for image compression. SINCO imposes structural consistency of the compressed images to the groundtruth by using a segmentation network to penalize the discrepancy of segmentation masks predicted from compressed images. We validate SINCO on brain MRI images by showing that it can achieve better performance than some recent INR methods.

Deep model-based architectures (DMBAs) are widely used in imaging inverse problems to integrate physical measurement models and learned image priors. Plug-and-play priors (PnP) and deep equilibrium models (DEQ) are two DMBA frameworks that have received significant attention. The key difference between the two is that the image prior in DEQ is trained by using a specific measurement model, while that in PnP is trained as a general image denoiser. This difference is behind a common assumption that PnP is more robust to changes in the measurement models compared to DEQ. This paper investigates the robustness of DEQ priors to changes in the measurement models. Our results on two imaging inverse problems suggest that DEQ priors trained under mismatched measurement models outperform image denoisers.

A better understanding of the mmWave channel coherence time will play a significant role in both communication and sensing. Traditional literatures consider that the channel coherence time in mmWave communication will be very short due to high frequency, which makes the channel estimation more difficult. However, we believe that the propagation characteristics of the mmWave channel will affect its coherence time, i.e., channel sparsity may limit its time variability. In this paper, we investigate the time-varying nature of mmWave channel by analyzing its Doppler spread based on the clustered sparsity model. After the derivation of Doppler spectrum DPS and the simulation analysis based on the two-cluster channel, we found that the maximum angle between clusters and the power distribution between clusters are two main factors limiting the Doppler spread. Finally, the realistic mmWave NYUSIM model is used for Doppler spread comparative analysis with Clarke’s model and find that Doppler spread of NYUSIM channel is far less than that estimated based on Clarke’s model. Thus, it is demonstrated that even if the mmWave frequency is high, the clustered sparsity characteristics makes its channel coherence time not as short as thought intuitively, which is meaningful for mmWave vehicular sensing and communication.

With the rapid development of the Internet, people pay more and more attention to network security and data privacy. Using the characteristics of SDN data and control separation, it is easy to embed a traffic detection model in edge devices to achieve abnormal traffic detection. However, although the traditional intrusion detection model can provide good recognition accuracy, it requires many labeled samples for model training. Not only is it challenging to obtain labeled samples, but it also brings privacy issues. This paper combines federated learning and anomaly-based CAE model in the SDN network and realizes intrusion detection on encrypted traffic under the premise of effectively protecting data privacy and reducing the workload of data labeling. Furthermore, we design an aggregation model selection algorithm based on loss and data volume evaluation, which reduces the overall training time of the federation and improves the model's accuracy.

Video captioning is a challenging task that captures different visual parts and describes them in sentences, for it requires visual and linguistic coherence. The attention mechanism in the current video captioning method learns to assign weight to each frame, promoting the decoder dynamically. This may not explicitly model the correlation and the temporal coherence of the visual features extracted in the sequence frames. To generate semantically coherent sentences, we propose a new Visual-aware Attention (VA) model, which concatenates dynamic changes of temporal sequence frames with the words at the previous moment, as the input of attention mechanism to extract sequence features. In addition, the prevalent approaches widely use the Teacher-forcing (TF) learning during training, where the next token is generated conditioned on the previous ground-truth tokens. The semantic information in the previously generated tokens is lost. Therefore, we design a Self-forcing (SF) stream that takes the semantic information in the probability distribution of the previous token as input to enhance the current token. The Dual-stream Decoder (DD) architecture unifies the TF and SF streams, generating sen-tences to promote the annotated captioning for both streams. Meanwhile, with the Dual-stream Decoder utilized, the ex-posure bias problem is alleviated, caused by the discrepancy between the training and testing in the TF learning. The effectiveness of the proposed Visual-aware Attention Dual-stream Decoder (VADD) is demonstrated through the result of ex-perimental studies on Microsoft video description (MSVD) corpus and MSR-Video to text (MSR-VTT) datasets.

Due to the characteristic of one-time parallel generation of a caption, non-autoregressive video captioning lacks strong dependencies between words. Although using guideline of scene-related visual words can promote caption generation, the semantic relations among visual words are barely explored, limiting the accurate representation. To this end, we propose a Dual-Scale Alignment-based transformer on Linguistic Skeleton Tags (DSA-LST), which alleviates the defect above in the form of visual words group (several words representing a video frame). Different groups represent different semantic dependencies by attention. We utilize linguistic skeleton tags (i.e., several groups) as sentence-level supervision for visual words sequence. For visual words group to accurately express a specific frame, we further design dual scales of visual-language bi-direction alignment to achieve internal relevance of the tags. Extensive experiments conducted on widely used datasets: MSVD and MSR-VTT demonstrate the effectiveness of our method when compared with existing approaches.

As Internet-of-Things (IoT) applications today utilize many diverse devices to collect information, Bluetooth low energy (BLE), featuring low power and low cost, is one of the most promising wireless solutions. To meet the requirements of diverse IoT applications, the neighbor discovery process (NDP) in BLE networks requires low cost and low latency, which is one of the most challenging tasks in supporting such a large number of BLE devices. Since the choice of BLE parameters is essential for achieving the required performance of BLE NDP, many performance analysis models have been proposed, aiming to provide guidance for the parameter configuration in IoT applications. This article reviews and studies the BLE NDP models and BLE performance analysis models proposed over the period 2012-2020, considering the advantages and constraints in utilizing these models in IoT. The performance analysis models are divided into two categories: 1) probabilistic models and 2) Chinese reminder theory-based models. The model design, performance metrics, deployment constraints, analysis results, and use cases are discussed for research, development, and applications.

We investigate an energy cost minimization problem for a smart home in the absence of a building thermal dynamics model with the consideration of a comfortable temperature range. Due to the existence of model uncertainty, parameter uncertainty (e.g., renewable generation output, nonshiftable power demand, outdoor temperature, and electricity price), and temporally coupled operational constraints, it is very challenging to design an optimal energy management algorithm for scheduling heating, ventilation, and air conditioning systems and energy storage systems in the smart home. To address the challenge, we first formulate the above problem as a Markov decision process, and then propose an energy management algorithm based on deep deterministic policy gradients. It is worth mentioning that the proposed algorithm does not require the prior knowledge of uncertain parameters and building the thermal dynamics model. The simulation results based on real-world traces demonstrate the effectiveness and robustness of the proposed algorithm.

The hop‐based localisation algorithm uses hop‐by‐hop propagation to establish node‐to‐anchor distance estimation, which does not require costly and complicated ranging hardware. This helps boost system performance, while minimising the cost of localising the nodes within the network. However, the application of hop‐based localisation algorithms is restricted due to their dramatic accuracy degradation in irregular network, which is mainly caused by the large error of distance estimation. The authors find that the error variance of the estimated distance increases as the hop count increases, i.e. there is a heteroscedasticity problem in the distance estimation process, which will affect the location estimation. In this study, by exploring the error during the location estimation, they aim to find and employ the optimal weighted function to improve localisation accuracy. A geometric constraint algorithm is also devised to correct the incorrectly estimated location by mitigating the adverse effects from flip ambiguity. By combining the optimal weighted function and the geometric constraint algorithm, a novel hop‐based localisation algorithm is proposed in this study. Both the theoretical analysis and experimental results show that the proposed method has not only maintained the economic characteristics of hop‐based localisation, but also has the high localisation accuracy where it can be adapted to various networks with different node distributions.

Attribute-based signature (ABS) is a promising cryptographic primitive. It allows the signer to generate a signature with attributes satisfying the predicate without leaking more information, so as to provide message authenticity in an anonymous manner. However, drawbacks concerning security and efficiency hinder its applications for authentication in mobile platforms. Here, we present F2P-ABS, an escrow-free and pairing-free attribute-based signature supporting perfect signer privacy for mobile anonymous authentication. To enhance its adaptiveness to mobile platforms, a novel key extraction is proposed so that the key escrow problem is mitigated over the single authority setting. It also helps to remarkably reduce the size of the signing key. Different from existing schemes, F2P-ABS is free from pairing operations. It performs no pairing operation for verification. Without the loss of security, we prove that F2P-ABS achieves signer privacy in perfect sense. It is also proven to guarantee existential unforgeability under corrupted and adaptive chosen predicate and message attack, which is securer than existing schemes.

Multi-view deep neural network is perhaps the most successful approach in 3D shape classification. However, the fusion of multi-view features based on max or average pooling lacks a view selection mechanism, limiting its application in, e.g., multi-view active object recognition by a robot. This paper presents VERAM, a view-enhanced recurrent attention model capable of actively selecting a sequence of views for highly accurate 3D shape classification. VERAM addresses an important issue commonly found in existing attention-based models, i.e., the unbalanced training of the subnetworks corresponding to next view estimation and shape classification. The classification subnetwork is easily overfitted while the view estimation one is usually poorly trained, leading to a suboptimal classification performance. This is surmounted by three essential view-enhancement strategies: 1) enhancing the information flow of gradient backpropagation for the view estimation subnetwork, 2) devising a highly informative reward function for the reinforcement training of view estimation and 3) formulating a novel loss function that explicitly circumvents view duplication. Taking grayscale image as input and AlexNet as CNN architecture, VERAM with 9 views achieves instance-level and class-level accuracy of 95.5 and 95.3 percent on ModelNet10, 93.7 and 92.1 percent on ModelNet40, both are the state-of-the-art performance under the same number of views.

Arterial traffic is the artery of urban transport and loads huge traffic pressure. In order to alleviate its traffic pressure effectively, a coordinated arterial traffic type-2 fuzzy logic control (FLC) method is proposed. First, arterial traffic flow model and evaluation index model are set up, in which the turning vehicles and lane length are given full consideration. The traditional queue spillover phenomenon in the traffic models can be prevented here. Second, aiming at the coordination and dynamic uncertainty problem in arterial traffic, a coordinated arterial traffic type-2 fuzzy coordination control method is put forward. It consists of two-layer type-2 fuzzy controller, the basic control layer and the coordination layer. The former allocates green time according to the traffic situation of each intersection, while the latter adjusts each intersection's green time on basis of the vehicles between the intersection and the downstream intersections for the purpose of enlarging green wave band. Finally, in order to configure the high-dimensional complex parameters of the coordinated two-layer type-2 FLC effectively, the parameters of membership function and the rules of the two controllers are optimized alternately by gravitational search algorithm. The simulation results verify the effectiveness of the proposed method from several aspects.

Wireless sensor networks (WSNs) provide a promising approach to monitor the physical environments, to prolong the network lifetime by exploiting the mutual correlation of sensor readings has become a research focus. In this paper, we propose a hierarchical network framework and adaptive threshold compression scheme to reduce the amount of information transmissions and alleviate the network congestion by exploring the spatial correlation among signals. The adaptive spatial compression scheme can obtain higher reconstruction precision by selectively discarding the less significant elements. Meanwhile, the compression ratio varies with the correlation among signals and adaptive threshold, so our scheme is adaptive to various deployed environments. Finally, the simulation results confirm that the proposed scheme achieves higher reconstruction precision and compression gain as compared with other spatial compression scheme.

Due to the limited energy of sensors and the difficulties in battery replacement in DTMSN (Delay Tolerant Mobile Sensor Network), unbalanced energy consumption will exhaust the batteries of active sensors soon, which can significantly reduce the network lifetime. Fortunately, this problem can be solved through multi-hop relay transmission where those energy-aware sensors with the maximum remaining energy will be selected to forward packets. However, one of the major challenges of multi-hop relay transmission in DTMSN is how to schedule these mobile sensors travelling paths in an energy-balanced way so that their overall lifetime is maximized. In this paper, an energy-balanced multi-hop relay transmission scheme based on RVNS (Reduced Variable Neighborhood Search) in DTMSN is proposed. Firstly, several parameters are designed to calculate the remaining energy of each sensor. Then RVNS is applied to obtain the global optimal solution. RVNS is implemented to select a sensor node with maximum remaining energy as the next hop relay, delivering packets to destination through multi-hop relay transmission. Simulation results demonstrate that in a delay-tolerant condition, the proposed scheme significantly balances the energy consumption and improves the packet delivery ratio.

Since the temporal and spatial correlations of sensor readings are existent in wireless sensor networks (WSNs), this paper develops a clustered spatio-temporal compression scheme by integrating network coding (NC) and compressed sensing (CS) for correlated data. The proper selections of NC coefficients and measurement matrix are designed for this scheme. This design guarantees the reconstruction of clustered compression data successfully with an overwhelming probability and unifies the operations of NC and CS into real field successfully. Moreover, in contrast to other spatio-temporal schemes with the same computational complexity, the proposed scheme possesses lower reconstruction error by employing the independent encoding in each sensor node (including the cluster head nodes) and joint decoding in sink node. At the same time it has lower computational complexity as compared with JSM-based spatio-temporal scheme by exploiting the temporal and spatial correlations of original sensing data step by step. Finally, the simulation results verify that the clustered spatio-temporal compression scheme outperforms the other two compression schemes significantly in terms of recovery error and compression gain.