Dynamic link prediction plays a crucial role in diverse applications including social network analysis, communication forecasting, and financial modeling. While recent Transformer-based approaches have demonstrated promising results in temporal graph learning, their performance remains limited when capturing complex multi-scale temporal dynamics. In this paper, we propose TFWaveFormer, a novel Transformer architecture that integrates temporal-frequency analysis with multi-resolution wavelet decomposition to enhance dynamic link prediction. Our framework comprises three key components: (i) a temporal-frequency coordination mechanism that jointly models temporal and spectral representations, (ii) a learnable multi-resolution wavelet decomposition module that adaptively extracts multi-scale temporal patterns through parallel convolutions, replacing traditional iterative wavelet transforms, and (iii) a hybrid Transformer module that effectively fuses local wavelet features with global temporal dependencies. Extensive experiments on benchmark datasets demonstrate that TFWaveFormer achieves state-of-the-art performance, outperforming existing Transformer-based and hybrid models by significant margins across multiple metrics. The superior performance of TFWaveFormer validates the effectiveness of combining temporal-frequency analysis with wavelet decomposition in capturing complex temporal dynamics for dynamic link prediction tasks. The code is available at https://github.com/SEUFHTong/TFWaveFormer.

Causal structure learning on event sequences is essential in critical systems such as communications, transportation, and industrial monitoring, where precise modeling of causal dependencies among event types significantly impacts tasks like root-cause alarm localization. Existing approaches often rely on Hawkes processes augmented with static network topology to eliminate the independent and identically distributed (i.i.d.) assumption, yet their dependence on predefined fixed structures and manually chosen kernels limits adaptability to dynamic network systems. The key challenge is to automatically infer accurate causal graphs from event sequences where both temporal evolution and network topology change over time, while accounting for nonlinear dependencies. To address this challenge, we propose Adaptive Neural Hawkes Processes (ANHP), a novel neural point process framework comprising three core modules: (1) Adaptive Event-Graph Learning, which constructs dynamic topology directly from event embeddings; (2) Neural Topological Hawkes Process, which replaces traditional linear excitation kernels with neural parameterization to capture nonlinear, time-varying conditional intensities; and (3) Masked Sparse Causal Structure, which balances likelihood maximization and model complexity via kernel-parameters masking and Bayesian Information Criterion (BIC) penalization to suppress redundant edges. Experimental results on real-world communication network alarm datasets from Huawei Shennong Intelligent Maintenance and Operation Center (IMOC) and synthetic datasets demonstrate that ANHP significantly outperforms state-of-the-art methods in accuracy, robustness, and scalability. The code is available at https://github.com/ChngYJ/ANHP.

This paper proposes ABIGX (Adversarial fault reconstruction-Based Integrated Gradient eXplanation), a unified framework for explainable fault detection and classification (FDC). ABIGX builds on the foundational principles of established fault diagnosis methods, including contribution plots (CP) and reconstruction-based contribution (RBC), while extending their applicability to general FDC models and improving fault explanation results. Central to ABIGX is the Adversarial Fault Reconstruction (AFR) method, which rethinks fault reconstruction from the perspective of adversarial attacks, introducing a novel fault index applicable to both fault detection and classification tasks. In fault detection, we theoretically bridge ABIGX with conventional fault diagnosis methods by proving that CP and RBC are the linear specifications of ABIGX. For fault classification, we address the challenge of fault class smearing, an inherent issue that can obscure accurate explanations. We demonstrate that ABIGX effectively mitigates this issue, outperforming current gradient-based explanation methods. The experiments evaluate the explanations of FDC by quantitative metrics and intuitive illustrations. The results validate the generality and accuracy of AFR, and show that ABIGX provides more comprehensive and precise explanations across various FDC models, offering a significant improvement over existing methods.

Hard shoulder running (HSR) has emerged as a sustainable and cost-effective strategy for improving expressway capacity. To address the limitations of existing approaches in capturing short-term traffic fluctuations and the coarse granularity of conventional sensor data, a short-term congestion prediction–driven, scenario-adaptive dynamic HSR (D-HSR) control framework based on multi-segment expressway video surveillance data is proposed. Specifically, a YOLOv8-DeepSORT pipeline is employed to extract real-time traffic flow parameters from video streams. Acongestion warning model is then developed to define dynamic control thresholds for HSR activation. Two distinct traffic scenarios are considered: recurrent and incident-induced congestion. The corresponding HSR activation decisions are formulated as time series forecasting (TSF) and traffic condition assessment (TCA) tasks, respectively. To enhance temporal modeling performance, S-Mamba is introduced as a high-capacity deep sequence model that enables more responsive and accurate traffic state predictions. The proposed strategy is implemented and evaluated on a calibrated simulation of the G25 section of the Changchun to Shenzhen expressway. Compared with the next-best model and rule-based baselines, the proposed method achieves a 1.95% and 31.13% reduction in average fuel consumption and a 4.06% and 58.53% reduction in average travel time, respectively. The results validate the effectiveness of the proposed strategy in facilitating intelligent and D-HSR operations for congestion mitigation.

Causal discovery aims to learn causal relationships between variables from targeted data, making it a fundamental task in machine learning. However, causal discovery algorithms often rely on unverifiable causal assumptions, which are usually difficult to satisfy in real-world data, thereby limiting the broad application of causal discovery in practical scenarios. Inspired by these considerations, this work extensively benchmarks the empirical performance of various mainstream causal discovery algorithms, which assume i.i.d. data, under eight model assumption violations. Our experimental results show that differentiable causal discovery methods exhibit robustness under the metrics of Structural Hamming Distance and Structural Intervention Distance of the inferred graphs in commonly used challenging scenarios, except for scale variation. We also provide the theoretical explanations for the performance of differentiable causal discovery methods. Finally, our work aims to comprehensively benchmark the performance of recent differentiable causal discovery methods under model assumption violations, and provide the standard for reasonable evaluation of causal discovery, as well as to further promote its application in real-world scenarios.

Spatio-temporal prediction is a pivotal task with broad applications in traffic management, climate monitoring, energy scheduling, etc. However, existing methodologies often struggle to balance model expressiveness and computational efficiency, especially when scaling to large real-world datasets. To tackle these challenges, we propose STH-SepNet (Spatio-Temporal Hypergraph Separation Networks), a novel framework that decouples temporal and spatial modeling to enhance both efficiency and precision. Therein, the temporal dimension is modeled using lightweight large language models, which effectively capture low-rank temporal dynamics. Concurrently, the spatial dimension is addressed through an adaptive hypergraph neural network, which dynamically constructs hyperedges to model intricate, higher-order interactions. A carefully designed gating mechanism is integrated to seamlessly fuse temporal and spatial representations. By leveraging the fundamental principles of low-rank temporal dynamics and spatial interactions, STH-SepNet offers a pragmatic and scalable solution for spatio-temporal prediction in real-world applications. Extensive experiments on large-scale real-world datasets across multiple benchmarks demonstrate the effectiveness of STH-SepNet in boosting predictive performance while maintaining computational efficiency. This work may provide a promising lightweight framework for spatio-temporal prediction, aiming to reduce computational demands and while enhancing predictive performance. Our code is avaliable at https://github.com/SEU-WENJIA/ST-SepNet-Lightweight-LLMs-Meet-Adaptive-Hypergraphs.

In Healthcare Internet of Things Networks (HIoTNs), safeguarding the sensitivity of patients’ electric healthcare records (EHRs) is imperative, necessitating effective privacy protection while ensuring ample data for training. Decentralized federated learning (DFL), as a peer-to-peer distributed machine learning architecture, serves a pivotal role in preserving the data privacy of EHRs. However, two primary challenges in applying DFL to HIoTNs include safeguarding the privacy of EHRs at individual hospitals and optimizing communication resources among hospitals. This article proposes a three-layer healthcare framework comprising a physical layer, a communication layer and an edge-encrypted layer to segregate hospital local data from interaction models. Moreover, a novel doubly accelerated DFL algorithm (AccDFL) is introduced to integrate DFL training and information interaction into the communication layer of HIoTNs. By employing the heavy-ball and Nesterov methods, AccDFL achieves double acceleration, ensuring both$\epsilon _{i}$-differential privacy (DP) and linear convergence. The theoretical analysis of AccDFL delves into the interaction among distributed gradient tracking (DGT), DP, and accelerated mechanisms, presenting comprehensive convergence and privacy analyses to overcome the exponential increase in parameters brought by the acceleration and DP mechanisms. Experimental results with realistic non-IID grayscale and color medical datasets of different disease types affirm the significant advantages of AccDFL over other algorithms in terms of accuracy and communication efficiency.

Time-series prediction has drawn considerable attention during the past decades fueled by the emerging advances of deep learning methods. However, most neural network based methods fail in extracting the hidden mechanism of the targeted physical system. To overcome these shortcomings, an interpretable sparse system identification method without any prior knowledge is proposed in this study. This method adopts the Fourier transform to reduces the irrelevant items in the dictionary matrix, instead of indiscriminate usage of polynomial functions in most system identification methods. It shows an visible system representation and greatly reduces computing cost. With the adoption of$l_{1}$norm in regularizing the parameter matrix, a sparse description of the system model can be achieved. Moreover, three data sets including the water conservancy data, global temperature data and financial data are used to test the performance of the proposed method. Although no prior knowledge was known about the physical background, experimental results show that our method can achieve long-term prediction regardless of the noise and incompleteness in the original data more accurately than the widely-used baseline data-driven methods. This study may provide some insight into time-series prediction investigations, and suggests that a white-box system identification method may extract the easily overlooked yet inherent periodical features and may beat neural-network based black-box methods on long-term prediction tasks.

Root cause analysis (RCA) is a crucial technique in network systems for uncovering the abnormal nodes that lead to the network alarm flood. Within private cloud network systems, the calling chains and topologies among entities, such as hosts, routes, and services, are always incomplete due to nonstandardized management. Existing topology-free RCA techniques, which rely on the casual discovery, are inapplicable when the scale of the network system is extremely large or the number of triggered alarms is sparse. This article proposes NetEventCause (NEC), an event-driven, unsupervised, and nonintrusive RCA algorithm for large network systems, where the network topology is unknown. NEC learns from historical alarm events to model the occurrences of various alarm types using a multivariate neural temporal point process (TPP). Based on the conditional intensity predicted by the learned TPP, NEC can identify the root alarms from a cascade of alarm events and locate the causal alarms of derivative alarms using the attribution method. The experimental section evaluates the NEC using both a synthetic event dataset and a large real-world dataset. The real-world dataset is exported from the Huawei Shennong Intelligent Maintenance and Operation Center (IMOC), a platform deployed at one of China's largest airports and manages over 200000 entities. Results obtained from the two datasets demonstrate that NEC outperforms most state of the art (SOTA) TPP models in modeling alarm events and surpasses general RCA methods in terms of identifying root alarms and recovering transmission chains of anomalies.

With the continuous advancement of neural network methodologies, time series prediction has attracted substantial interest over the past decades. Nonetheless, the interpretability of neural networks is insufficient and the utilization of deep learning techniques for prediction necessitates significant computational expenditures, rendering its application arduous in numerous scenarios. In order to tackle this challenge, an interpretable sparse system identification method which does not require a time-consuming training through back-propagation is proposed in this study. This method integrates advantages from both knowledge-based and data-driven approaches, and constructs dictionary functions by leveraging Fourier basis and taking into account both the long-term trends and the short-term fluctuations behind data. By using the $l_1$ norm for sparse optimization, prediction results can be gained with an explicit sparse expression function and an extremely high accuracy. The performance evaluation of the proposed method is conducted on comprehensive benchmark datasets, including ETT, Exchange, and ILI. Results reveal that our proposed method attains a significant overall improvement of more than 20\% in accordance with the most recent state-of-the-art deep learning methodologies. Additionally, our method demonstrates the efficient training capability on only CPUs. Therefore, this study may shed some light onto the realm of time series reconstruction and prediction.

Multimodal data fusion analysis is essential to model the uncertainty of environment awareness in digital industry. However, due to communication failure and cyberattack, the sampled time-series data often have the issue of data missing. In some extreme cases, part of units are unobservable for a long time, which results in complete data missing (CDM). To impute missing data, many models have been proposed. However, they cannot address the CDM issue, because no observation data of the unobservable units can be obtained in this case. Thus, to address the CDM issue, a novel cross-modal generative adversarial network (CM-GAN) is proposed in this article. It combines the cross-modal data fusion technique and the deep adversarial generation technique to construct a cross-modal data generator. This generator can generate long-term time-series data from widely existing spatio-temporal modal data in modern industrial system, and then impute missing value by replacing them with generated data. To test the performance of CM-GAN, extensive experiments are conducted on photovoltaic (PV) power output dataset. Compared with other baseline models, the performance of CM-GAN is generally better and reaches the state-of-the-art level. Moreover, sufficient ablation studies are conducted to present the contribution of the cross-modal data fusion technique and show the reasonability of parameter settings of CM-GAN. Apart from this, some prediction experiments are also conducted. The results show that the PV data recovered by CM-GAN can provide more predictability information for improving the prediction accuracy of deep learning model.

Dynamic resource allocation problem (DRAP) with unknown cost functions and unknown resource transition functions is studied in this article. The goal of the agents is to minimize the sum of cost functions over given time periods in a distributed way, that is, by only exchanging information with their neighboring agents. First, we propose a distributed Q -learning algorithm for DRAP with unknown cost functions and unknown resource transition functions under discrete local feasibility constraints (DLFCs). It is theoretically proved that the joint policy of agents produced by the distributed Q -learning algorithm can always provide a feasible allocation (FA), that is, satisfying the constraints at each time period. Then, we also study the DRAP with unknown cost functions and unknown resource transition functions under continuous local feasibility constraints (CLFCs), where a novel distributed Q -learning algorithm is proposed based on function approximation and distributed optimization. It should be noted that the update rule of the local policy of each agent can also ensure that the joint policy of agents is an FA at each time period. Such property is of vital importance to execute the ε -greedy policy during the whole training process. Finally, simulations are presented to demonstrate the effectiveness of the proposed algorithms.

This paper investigates the spatiotemporal characteristics and predictability of the emerging modern traffic behavior, ridesourcing. We collect a comprehensive data set of Didi ridesourcing cars on a large geographical scale of a capital city in China, including both the temporal order information and the GPS-recorded spatial trajectories. To extract the features of this kind of traffic behavior, we construct a large-scale network by considering every traffic flow of the orders. Therein, a driver consecutively visiting different regions of the city connects the relationship of these sites. The weighted ridesourcing network shows a consistency of the distribution of trip orders and the Clark model for population distribution. The network also has spatial and temporal features with power laws, sometimes with exponential truncations and log-normal distributions. Furthermore, we propose a general analytical method to quantify the predictability of this kind of behavior by calculating the entropy at a collective level, which can be extended to quantify other traffic behaviors. Finally, by considering the traffic congestion factor, we propose a better neural network based model for predicting dwelling time of the ridesourcing behavior. We suggest that the traffic behavior of ridesourcing cars indicates specific non-Markovian characteristics, which can be systematically analyzed from the viewpoint of network sciences.

Formation control finds broad applications in numerous fields, such as cooperative detection, surveillance, transportation, and disaster rescue. In this article, aiming at hunting a moving target, a two-stage surrounding control algorithm is proposed for linear multiagent systems subject to input saturations. An adaptive distributed observer is developed for each agent to reconstruct the target’s position. With the assistance of the algebraic graph theory and low gain feedback technique, distributed controllers are designed to drive the agents to encircle the moving target with a fixed radius and evenly distributed phase angles. Finally, both numerical simulations and experiments are conducted to verify the effectiveness of the proposed control algorithm.

This paper presents a fuzzy-model-based method for solving the consensus problem of a class of nonlinear multi-agent systems (MASs) with input saturation. Since each agent has nonlinear dynamics, the system is not asymptotically null controllable with bounded controls (ANCBC). Therefore, the widely-used low-gain feedback method for designing consensus protocols of MASs with input saturation can no longer work. To this end, the Takagi-Sugeno (T-S) fuzzy model is adopted to formulate the error dynamics of those nonlinear follower agents with input saturation as well as a leader with time-varying states. Accordingly, by using the properties of convex hull, a set of invariance condition in the format of linear matrix inequality (LMI) is designed. Furthermore, by enlarging the shape reference set, the estimation of the attraction domain can be obtained. Simultaneously, by viewing the control gain as an extra free parameter in the LMI optimization procedure, the leader-follower consensus algorithm is proposed, which guarantees that all followers with input saturation can track the leader, and they can asymptotically reach consensus. Finally, numerical experiments validate the effectiveness of the proposed anti-saturation consensus algorithm.