Recent studies have demonstrated that hypergraph neural networks (HGNNs) are susceptible to adversarial attacks. However, existing methods rely on the specific information mechanisms of target HGNNs, overlooking the common vulnerability caused by the significant differences in hyperedge pivotality along aggregation paths in most HGNNs, thereby limiting the transferability and effectiveness of attacks. In this paper, we present a novel framework, i.e., Transferable Hypergraph Attack via Injecting Nodes into Pivotal Hyperedges (TH-Attack), to address these limitations. Specifically, we design a hyperedge recognizer via pivotality assessment to obtain pivotal hyperedges within the aggregation paths of HGNNs. Furthermore, we introduce a feature inverter based on pivotal hyperedges, which generates malicious nodes by maximizing the semantic divergence between the generated features and the pivotal hyperedges features. Lastly, by injecting these malicious nodes into the pivotal hyperedges, TH-Attack improves the transferability and effectiveness of attacks. Extensive experiments are conducted on six authentic datasets to validate the effectiveness of TH-Attack and the corresponding superiority to state-of-the-art methods.

Aiming at the problems of redundant information accumulation, low computational efficiency, and fuzzy feature allocation in multiscale time-series prediction of traditional deep echo state network (DeepESN), this article proposed an interlayer sparse compression-based DeepESN model (ICS-DESN). The model uses the sparse sampling technology of deep fusion compressive sensing and the hierarchical dynamic feature extraction mechanism of DeepESN, introduces the adaptive compressed sampling module between the layers, and uses the Gaussian observation matrix to reduce the dimension of the high-dimensional state, which effectively inhibits the stacking of redundant information in the deep network, and explicitly allocates the multiscale temporal features. Through theoretical analysis, it is proven that ICS-DESN satisfies the stability condition of the echo state property (ESP) by constraining the weighted spectral radius of the reservoir. In the experiment, we used multiscenario time-series datasets, such as logistic chaotic systems, Lorenz attractors, sunspot data, NASDAQ stock index, ETTh1 dataset, and weather dataset to validate the effectiveness of the model. The results showed that compared with traditional comparison models, ICS-DESN significantly reduced prediction errors [mean squared error (MSE) and mean absolute error (MAE)], demonstrating higher computational efficiency and robustness. This research provides an efficient theoretical framework for complex time-series modeling and has potential application value in resource-constrained scenarios, such as edge computing.

Mild cognitive impairment (MCI), an early stage of Alzheimer's disease, is marked by subtle cognitive decline. While graph neural networks (GNNs) have shown promise in modeling neuroimaging-based brain networks for MCI detection, most treat the brain as a homogeneous graph, ignoring hemispheric structural and functional heterogeneity. To address this limitation, we propose a Hemispheric Heterogeneity-aware Graph Neural Network (HemiHeter-GNN) to explicitly model hemispheric structural and functional heterogeneity for improved MCI detection. Multimodal brain networks are constructed by integrating diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI), and decomposed into left-, right, and inter-hemispheric subnetworks to capture hemispherespecific patterns. Each subnetwork is encoded by a dedicated graph encoder and refined by heat-kernel diffusion for topological regularity, while a cross-subnetwork attention mechanism then fuses them into a unified whole-brain embedding. Experiments on the ADNI dataset show that HemiHeter-GNN consistently outperforms state-of-the-art baselines, demonstrating the effectiveness of modeling hemispheric asymmetry for MCI detection.

Hypergraph-based models have shown great potential in capturing high-order interactions within brain functional networks for disease identification. However, most hypergraph neural networks (HGNNs) depend on static and heuristically constructed hypergraphs, which fail to model the brain's dynamic connectivity patterns. In this paper, we propose a Cluster-Aware Hypergraph Network that jointly learns functional clusters and adaptive high-order connections, enhanced by contrastive learning to preserve topological consistency. Comprehensive experiments conducted on three real-world datasets demonstrate the effectiveness of our proposed methods.

Accurate electroencephalogram(EEG) annotation is essential for seizure detection but costly and error-prone, which can affect subsequent tasks. Moreover, the brain is a non-Euclidean topological structure, which contains spatial information for seizure detection. Based on these, this paper proposes a semi- supervised model based on Gaussian mixture variational autoencoder with graph representation, named GGMVAE. Firstly, for unlabeled EEG signals, we construct an adjacency matrix via Pearson correlation between channels. Then, matrix and EEG features are fed into a Gaussian Mixture VAE to learn the temporal and spatiall features through unsupervised training. Finally, the pre-trained encoder extracts low-dimensional features from partially labeled data for classification. The method is evaluated on the epilepsy dataset at the University of Helsinki, achieving the accuracy of 97.27 %, precision of 96.17 %, recall of 98.41 %, and F1-score of 97.28 %. The results indicate that this semi-supervised model can effectively learn the temporal and spatial features of EEG signals, improving the performance of seizure detection..

Hypergraphs provide excellent modeling ability for brain disorder identification, especially in capturing high-order interactions among regions of interest (ROIs). Nevertheless, existing methods overlook the impact of directional hyperedges learning on the brain network, leading to wasteful functional connectivity and poor identification performance. To address the above issue, this paper proposes a novel Brain Disorder Identification method via Directed Hypergraph Networks (D-HyperNet). Specifically, our methodology employs an Effective Network Construction module to capture causal dependencies and infer directional functional connectivity among ROIs. Followed by the Flow-aware Feature Aggregation module, which designs a novel directed hypergraph encoder that directionally aggregates node features, effectively improving the accuracy and reliability of brain network representations. Additionally, we are integrating the proposed encoder into a contrastive learning program to obtain a more robust whole-brain representation. Extensive experiments demonstrate the efficacy of our D-HyperNet approach. The code is available at https://github.com/Jia-Weiming/D-HyperNet.

Analyzing brain network connectivity (BNC) using multimodal neuroimaging to identify neurodegenerative diseases has attracted increasing attention. However, current methods largely rely on node-centric graphs and assume structural or semantic alignment across modalities, limiting the capture of high-order interactions and modality-specific patterns critical for accurate disease identification. In this paper, we propose a novel Hyperedge-Centric Graph Learning Network (HCGLNet) to address these limitations. Specifically, we present a hyperedge-centric graph construction strategy (HGC) that represents each modality as a hyperedge-centric graph, explicitly modelling high-order connectivity unique to each modality. Moreover, we design a disentangled latent learning module (DLM) that factorize shared and specific representations, preserving modality-specific features from dilution while enabling the extraction of shared cross-modal representations. Finally, we develop a representation-aware routing (RAR) algorithm to adaptively fuse modality-specific and shared features based on learned weights, enhancing discriminability for downstream tasks. Experiments on three real-world datasets show that our HCGLNet abstraction reduces graph size by over 80%, lowers computation, and achieves state-of-the-art performance in neurodegenerative disease classification.

Population graphs are crucial for assessing clinical risk and enhancing the accuracy of Autism Spectrum Disorder (ASD) detection. Nevertheless, the current population graph construction overlooks the balance between biological signals and clinical manifestations, leading to relationship deviation within the population graph and poor detection performance. To address this challenge, we propose a novel approach for ASD Detection in Population Graphs (PopuDet) via Micro-macro Relationship Construction (MmRC) and Multi-feature Fusion (MF). Specifically, our method utilizes the MmRC module to construct a multi-scale population graph balancing the relationships between biological signals synchrony and clinical subtype groups. Subsequently, the MF module learns high-level graph representations at different scales for adaptive fusion to achieve precise ASD detection. Extensive experiments validate the efficiency of PopuDet, highlighting its superior performance over current state-of-the-art methods. Our source code is available at https://github.com/xuting99/PopuDet.

Understanding the communication and information processing mechanisms in pinning control is highly influenced by the connectivity structure of the network, particularly in higher-order networks. However, existing studies predominantly focus on static network structures and utilize complete error information, often neglecting underlying inter-node dependencies, thereby limiting the effectiveness of low-energy control in higher-order dynamic systems. To address these limitations, this paper proposes a novel pinning control scheme for the synchronization of inertial memristive neural networks (IMNNs) based on dynamic Granger causality analysis (DGCA). First, an interpretable IMNN model is constructed to explicitly characterize higher-order interactions without decomposing the system into first-order subsystems. Then, unlike traditional pinned-node selection strategies relying on static interaction rules, a causality-aware selection algorithm is developed using DGCA to dynamically identify influential nodes via time-series analysis, enhancing control efficiency under dynamic network conditions. Furthermore, a local information-based pinning controller is designed by leveraging local causal relationships and control influence regions extracted from DGCA, ensuring the stability of the synchronization error system. Theoretical guarantees are provided by deriving sufficient conditions for controller design based on Lyapunov stability theory. Finally, three numerical examples are presented to demonstrate the effectiveness and practicality of the proposed scheme.

The precise detection of mild cognitive impairment (MCI) is of significant importance in preventing the deterioration of patients in a timely manner. Although hypergraphs have enhanced performance by learning and analyzing brain networks, they often only depend on vector distances between features at a single scale to infer interactions. In this paper, we deal with a more arduous challenge, hypergraph modelling with synchronization between brain regions, and design a novel framework, i.e., A Multi-scale Hypergraph Network for MCI Detection via Synchronous and Attentive Fusion (MHSA), to tackle this challenge. Specifically, our approach employs the Phase-Locking Value (PLV) to calculate the phase synchronization relationship in the spectrum domain of regions of interest (ROIs) and designs a multi-scale feature fusion mechanism to integrate dynamic connectivity features of functional magnetic resonance imaging (fMRI) from both the temporal and spectrum domains. To evaluate and op-timize the direct contribution of each ROI to phase synchronization in the temporal domain, we structure the PLV coefficients dynamically adjust strategy, and the dynamic hypergraph is modelled based on a comprehensive temporal-spectrum fusion matrix. Experiments on the real-world dataset indicate the effectiveness of our strategy. The code is available at https://github.com/Jia-Weiming/MHSA.

This paper studies the problem of dynamic event-triggered (ET) output feedback (OF) control for switched nonlinear systems (SNSs) with uncertain output function. The considered system allows more general growth restriction. Firstly, to deal with the system uncertainties, a homogeneous observer embedded with a dynamic gain is put forward. Then, an adaptive control scheme under arbitrary switching is presented. It is worth emphasizing that the threshold parameters of the used ET mechanism can be dynamically adjusted. Besides, it is proven that all signals of the closed-loop system are bounded. The effectiveness of the presented method is verified by an illustrative example.

This article solves the exponential synchronization issue of memristor-based complex-valued neural networks (MCVNNs) with time-varying uncertainties via feedback control. Compared with the traditional control methods, a more practical and general control scheme with the available uncertain information of the parameters is newly developed for MCVNNs. Our approach considers the proposed neural networks as two dynamic real-valued systems. Then, the less conservative exponential synchronization criteria are proposed by incorporating the framework of the Lyapunov method and inequality techniques. Under the proposed algorithm, not only can the stability of MCVNNs be guaranteed but also the behavior of such a system is appropriate for image protection. Meanwhile, the sensitive measure of the encryption and decryption can be converted into synchronization error. When monitoring the secure mechanism as a whole, the influence of error feasible domain on image decryption is analyzed. Simulation examples are provided to verify the efficacy of the proposed synchronization criterion and the results of practical application on image protection.