查看更多>>摘要：Electroencephalographic (EEG) source imaging (ESI) aims to estimate brain activity locations and extents. ESI is crucial for studying brain functions and detecting epileptic foci. However, accurately reconstructing extended sources remains challenging due to high susceptibility of EEG signals to interference and the underdetermined nature of the ESI problem. In this study, we introduce a new ESI algorithm, Structured Sparse Source Imaging based on L-2p-norm (SSSI-L-2p), to estimate potential brain activities. SSSI-L-2p utilizes the mixed L-2p-norm (0<p<1) to enforce spatial-temporal constraints within a structured sparsity regularization framework. By leveraging the alternating direction method of multipliers (ADMM) and iteratively reweighted least squares (IRLS) algorithm, the challenging optimization problem of SSSI-L-2p can be effectively solved. We showcase the superior performance of SSSI-L-2p over benchmark ESI methods through numerical simulations and human clinical data. Our results demonstrate that sources reconstructed by SSSI-L-2p exhibit high spatial resolution and clear boundaries, highlighting its potential as a robust and effective ESI technique. Additionally, we have shared the source code of SSSI-L-2p at https://github.com/Mashirops/SSSI-L2p.git
查看更多>>摘要：This study addresses the privacy-preserving average consensus problem in second-order discrete multi-agent systems under strongly connected and balanced graphs. When both velocity and position states of each agent are measurable, a novel lightweight algorithm is proposed by introducing perturbation signals into the transmitted information. Specifically, the algorithm is divided into two stages. In the initial stage, each agent introduces perturbation signals into its initial position and velocity states during transmission to confound potential attackers. In the subsequent stage, the agents use a standard average consensus algorithm to update their states, ensuring accurate convergence to the average of the initial states. Additionally, further considering the scenario where the velocity state is unavailable for each agent, an improved edge-based perturbation algorithm is introduced. Both algorithms not only effectively prevent the internal honest-but-curious agents from accurately inferring the initial states of other agents, except in the specific case where the curious agent is the sole neighbor of the target agent, but also protect privacy from the external eavesdroppers. Lastly, several numerical examples are conducted to validate the effectiveness of the proposed theoretical approaches.
查看更多>>摘要：In this paper we show how tensor networks help in developing explainability of machine learning algorithms. Specifically, we develop an unsupervised clustering algorithm based on Matrix Product States (MPS) and apply it in the context of a real use-case of adversary-generated threat intelligence. Our investigation proves that MPS rival traditional deep learning models such as autoencoders and GANs in terms of performance, while providing much richer model interpretability. Our approach naturally facilitates the extraction of feature-wise probabilities, Von Neumann Entropy, and mutual information, offering a compelling narrative for classification of anomalies and fostering an unprecedented level of transparency and interpretability, something fundamental to understand the rationale behind artificial intelligence decisions.
查看更多>>摘要：Semi-supervised nonnegative matrix factorization (NMF) has been widely used in various clustering tasks due to its reliable performance. The key is how to use effectively a small amount of label information to obtain a more discriminative low-dimensional representation of data. In order to improve the clustering performance of semi-supervised NMF more effectively, this paper proposes a new semi-supervised NMF method, namely semi-supervised structured NMF for anchor graph embedding (AESSNMF). Specifically, AESSNMF uses three kinds of supervision information simultaneously, namely, pointwise constraints, pairwise constraints, and negative label information. Also, in order to handle mixed-sign data, AESSNMF uses a convex NMF form and only imposes nonnegative constraints on the coefficient matrix. AESSNMF constructs an anchor graph to embed the matrix factorization process, rather than performing the matrix factorization directly on the original data. We use the alternating iterative algorithm to optimize the objective function of AESSNMF. We also discuss the relationship between several related NMF based algorithms and AESSNMF. A large number of experimental results show that AESSNMF is superior to other related algorithms.
查看更多>>摘要：This paper investigates high-order nonlinear multi-agent systems with state constraints and input saturation. A novel control scheme incorporating Neural Networks and Barrier Lyapunov Functions is designed to achieve adaptive fixed-time consensus control. This innovative scheme effectively addresses the complexity explosion problem typical in traditional controller designs while ensuring that the closed-loop system remains within its constraints. During the design process, a first-order sliding mode differentiator was introduced, and compensations were made for filter errors to ensure stability and consistency within a fixed-time. Additionally, experiments using Matlab numerical simulations and the StarSim semi-physical simulation platform confirm that the proposed control scheme significantly surpasses traditional methods in efficiency and accuracy, validating its effectiveness and practicality for solving the consensus problem in high-order nonlinear multi-agent systems.
Yu, SudanChen, HuilingHeidari, Ali AsgharLiang, Guoxi...
1.1-1.26页
查看更多>>摘要：Feature selection is a critical process in machine learning and data mining that selects a subset of relevant features for model construction. This process helps reduce the dimensionality of the data, minimize computational cost, improve model performance, and enhance interpretability. An effective feature selection method can significantly impact the success of predictive modeling tasks. Recently, numerous algorithms have been developed to address the challenges associated with high-dimensional datasets. These algorithms range from traditional statistical methods to advanced metaheuristic methods. The RIME approach, enhanced through reinforcement learning, presents a viable answer to the optimization challenge. Through reinforcement learning that switches between the dispersed foraging and hard-RIME puncture mechanisms and comprehensive learning strategy, the multi-strategy ensembled RIME method, named QCLFRIME, reaches a more effective balance between local search and global exploration. This algorithm is validated on the IEEE CEC 2017 function test and compared with traditional and advanced algorithms. Subsequently, we develop a wrapper feature selection model based on the binary QCLFRIME model. Besides, the binary QCLFRIME-KNN classifier has excellent performance for the feature selection regarding fitness, error rate, the number of selected features and computational time on 14 publicly available high-dimensional datasets compared to the traditional and enhanced meta-heuristic algorithms. Relevant experimental findings highlight the proposed algorithm's outstanding efficacy, which surpasses most existing algorithms. Therefore, the proposed method is a valuable wrapper-mode feature selection tool.
查看更多>>摘要：This study proposes a novel, effective improved predictive function control based on a wavelet neural network (IPFC-WNN) for urban rail train tracking control. Specifically, the step function and Morlet wavelet function were chosen as the base function together, and an adaptive nonlinear online adjustment function of the softening factor was proposed based on the fuzzy satisfaction of system performance and optimisation factor. The maximum and minimum softening factors for a simple straight line can also be set appropriately by a wavelet neural network according to the actual situation. To effectively improve the control performance of the predictive function control algorithm for urban rail train tracking, calculation of additional resistance with a multiparticle model was adopted, and parameters for the adaptive nonlinear online softening factor adjustment function were set using a wavelet neural network to improve the comprehensive performance quality for urban rail train tracking control. Considering the scenario of urban rail train tracking control from Bayi Road to Yongan Four Seasons, which is located in the second-phase project of Dalian Urban Rail Transit Line 13, as the hardware-in-the-loop test object, the proposed IPFC-WNN and three improved control algorithms were used for comparative verification. The test results showed that the proposed IPFC-WNN can significantly improve the performance of the control system, and quality indicators such as energy saving, precise parking, punctuality, and comfort of the system were significantly improved. Hence, the good tracking control for train operation using the proposed IPFC-WNN was verified.
查看更多>>摘要：Accurate prediction of nonlinear systems in non-Gaussian noise environments has long been a significant challenge in the fields of statistical data analysis and time series modeling. To address this issue, this paper proposes an improved Cauchy Conjugate Gradient algorithm based on a kernel broad learning feature extraction strategy (Kernel Broad Learning Cauchy Conjugate Gradient, KBLCCG). This algorithm integrates kernel mapping with broad learning systems, forming a dual feature extraction mechanism that effectively captures the complex nonlinear structures of chaotic time series while preserving their inherent dynamic chaotic characteristics. The KBLCCG algorithm utilizes its robust feature extraction capabilities through the dual extraction mechanism of kernel mapping and broad learning systems, effectively capturing the intricate nonlinear structures present in time series data. The kernel broad learning strategy mitigates the phenomenon of kernel matrix size expansion during the iterative process, thereby reducing the computational burden and enhancing the algorithm's robustness. The Cauchy Conjugate Gradient method is employed to optimize the reduced-dimensional feature data, efficiently addressing the nonlinear prediction problem of the target sequence. Empirical analysis using simulation data and actual financial data (including the Lorenz system, Shanghai Composite Index, and CSI 300 Index) validates the performance of this method. Experimental results indicate that KBLCCG significantly outperforms existing adaptive filtering algorithms in terms of prediction accuracy, particularly demonstrating stronger generalization capabilities when dealing with complex chaotic systems. Compared to traditional methods, the kernel broad learning strategy markedly enhances the feature capturing and modeling effectiveness of chaotic time series, further validating the method's efficacy and robustness in nonlinear time series prediction. The KBLCCG algorithm not only exhibits superior predictive capabilities in complex non-Gaussian noise environments but also provides an innovative solution for handling the nonlinear and chaotic characteristics of time series prediction.
查看更多>>摘要：Infrared image super-resolution reconstruction is crucial for improving the quality of infrared images. The cutting-edge efforts often suffer from problems such as unnatural texture and fake details, undermining image realism. This study introduces a novel unified framework, which combines a multi-order degradation simulation model and region-aware discriminative learning with generative adversarial network, to train the optimal model for infrared super-resolution reconstruction. The multi-order degradation model is established to enhance the effectiveness of visual reconstruction by simulating different degradation instances in authentic infrared scenarios. The generator adopts residual-in-residual dense blocks to enhance detail preservation capability. The discriminator uses an encoder-decoder architecture for semantic discrimination. This enables simultaneous global and local assessments of images as real or fake, providing a more accurate discrimination of the generated image's authenticity. Additionally, to discriminate high-frequency artifacts and authentic details in infrared images with intricate textures, a region-aware discriminative learning strategy is introduced. Furthermore, a hybrid loss function is developed, integrating local loss, adversarial loss, pixel-wise and perceptual loss together for advanced adversarial training, making the reconstructed images more realistic and natural. The method's efficacy is demonstrated on simulated and real-world infrared image datasets, with comparative analysis showing significant improvements. Our model outperforms the state-of-the-art alternative solutions on multiple benchmarks.
查看更多>>摘要：Aspect Sentiment Triplet Extraction (ASTE) has emerged as a pivotal task in sentiment analysis, focusing on extracting the aspect terms along with the corresponding opinion terms and the expressed sentiments. Recently, generative models have achieved significant success in ASTE task. However, existing generative approaches fail to further model the specific relations within the context for ASTE at the encoding phase, making it difficult to establish the nuanced connections between aspect and opinion terms. Additionally, these approaches rely on simple structured templates at the decoding phase to pair aspect terms with opinion terms, which fails to provide effective relation information for the decoding process. To address the aforementioned issues, we propose ReHyGen, a novel relational hypergraph enhanced framework designed to enhance the relational modeling capabilities of generative ASTE models during both the encoding and decoding phases. Specifically, ReHyGen comprises two core components: the Relational Hypergraph Enhanced Module (RHEM) and the Relational Prompt Module (RPM). RHEM leverages the hypergraph attention network and auxiliary relation classification to capture high-order word interactions and boundary-sensitive word pair relations. RPM incorporates relational information into the decoding phase by providing relation-aware prompts, guiding the generation of more accurate target sequences. Extensive experiments on benchmark datasets demonstrate that our proposed framework significantly improve the performance of generative ASTE models.
NETL
NSTL
Elsevier