Article
Mathematics, Interdisciplinary Applications
M. Hymavathi, G. Muhiuddin, M. Syed Ali, Jehad F. Al-Amri, Nallappan Gunasekaran, R. Vadivel
Summary: This paper investigates the global exponential stability of fractional order complex-valued neural networks with leakage delay and mixed time varying delays. Sufficient conditions for global exponential stability are established by constructing a proper Lyapunov-functional. The stability conditions are expressed in terms of linear matrix inequalities and the effectiveness of the obtained results is illustrated through two numerical examples.
FRACTAL AND FRACTIONAL
(2022)
Article
Automation & Control Systems
Xiaonan Liu, Minghui Song, Yonggui Kao
Summary: This paper investigates the synchronization between two hyperbolic coupled networks (HCNs) with time-varying delays using aperiodically intermittent pinning control (AIPC). Sufficient criteria for HCNs with internal delays only and with hybrid delays are obtained based on a Lyapunov function with a piecewise continuous function. It is found that HCNs with hybrid delays have a slower convergence speed compared to those with internal delays only. Additionally, two simulation results are presented to validate the theoretical findings.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Yin Sheng, Tingwen Huang, Zhigang Zeng, Xiangshui Miao
Summary: This article investigates the Lagrange exponential stability and the Lyapunov exponential stability of memristive neural networks with discrete and distributed time-varying delays. The study uses inequality techniques, theories of the M-matrix, and the comparison strategy to consider the stability of the networks, providing less conservative methods for analyzing Lyapunov stability.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Kai Wu, Jigui Jian
Summary: This article focuses on the global robust exponential dissipativity (GRED) of uncertain second-order BAM neural networks with mixed time-varying delays. New differential inequalities and Lyapunov-Krasovskii functionals are established to present new GRED criteria in the form of linear matrix inequalities. The correctness of the theoretical results is verified through simulation experiments.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Information Systems
Imran Ghous, Jian Lu, Zhaoxia Duan
Summary: This work investigates the stability and stabilization problems of memristive neural networks (MNNs) considering time-varying delay and external disturbance. The MNNs are transformed into a tractable model by defining logical switched functions. A new Lyapunov-Krasovskii functional is proposed to study the exponential stability (ES) problem of the transformed MNNs model. The design scheme of a state feedback controller is devised to ensure the stability of the overall closed-loop system. The efficacy of the proposed results is demonstrated through suitable examples.
INFORMATION SCIENCES
(2022)
Article
Mathematics, Applied
Qinghua Zhou, Li Wan, Hongshan Wang, Hongbo Fu, Qunjiao Zhang
Summary: Due to the inability to convert Cohen-Grossberg neural networks with multiple time-varying delays and distributed delays into vector-matrix forms, the stability results and conditions in the linear matrix inequality forms are relatively few and missing. This paper addresses the issue by investigating the exponential stability of the networks and providing sufficient conditions in the linear matrix inequality forms. Two examples are used to demonstrate the effectiveness of the theoretical results.
Article
Computer Science, Artificial Intelligence
Qiao Chen, Xinge Liu, Xuemei Li
Summary: This paper presents an improved approach for the exponential stability of neural networks with time-varying delay, establishing a less conservative stability criterion by combining various inequalities. The effectiveness and benefits of the proposed method are illustrated through several numerical examples.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Mathematics, Applied
El Abed Assali
Summary: This paper investigates the global exponential stability of a class of Clifford-valued recurrent neural networks with time-varying delays and distributed delays. The existence of equilibrium point for Clifford-valued recurrent neural networks is established based on Brouwer's fixed point theorem. By using inequality technique and the method of the Clifford-valued variation parameter, novel assertions are given to ensure the global exponential stability of the model, which complement some previous works. The effectiveness of this approach is illustrated with a numerical example.
COMPUTATIONAL & APPLIED MATHEMATICS
(2023)
Article
Automation & Control Systems
Shuhao Cao, Xian Zhang, Tianqiu Yu, Xiaona Yang
Summary: This article investigates the global h-stability for differential positive systems with multiple discrete time-varying delays and constant distributed delays. By proposing a direct analysis method, a sufficient condition for the global h-stability is obtained and represented in simple inequality form for easy handling.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2022)
Article
Computer Science, Software Engineering
Manchun Tan, Zhiqiang Song, Yunfeng Liu, Zhong Li
Summary: The problem of global robust exponential stability of delayed interval Cohen-Grossberg neural networks is addressed by establishing suitable Lyapunov functional and applying the linear matrix inequality (LMI) technique. The derived conditions ensure the existence, uniqueness, and robust exponential stability of equilibrium points. Numerical examples demonstrate that the new results are less restrictive and conservative compared to existing ones in the literature.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Song Zhu, Jiahui Zhang, Xiaoyang Liu, Mouquan Shen, Shiping Wen, Chaoxu Mu
Summary: This article analyzes the multistability and robustness of competitive neural networks with time-varying delays. Sufficient conditions are proposed based on the geometry of activation functions to determine the existence of equilibrium points and their stability. The conclusions proposed in this article are easy to verify and enrich the existing theories.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Fei Wei, Guici Chen, Wenbo Wang
Summary: This paper investigates the finite-time stability problem of memristor-based inertial neural networks (MINNs) with time-varying delays using the interval matrix approach. The study converts MINNs to systems with uncertain terms and employs dimension reduction to simplify the differential systems. Two types of delayed feedback controllers are designed to achieve finite time stabilization, with the stability criterion deduced by linear matrix inequalities. The theoretical results and proposed method are validated through two examples.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Mathematics, Applied
Jie Pan, Zhaoya Pan
Summary: This paper focuses on the robust stability of uncertain parameter quaternionic neural networks (QNNs) with both time-varying delays and infinite distributed delays. A derivative formula of quaternionic function's norm is established to obtain algebraic standards for global robust exponential stability. The whole quaternionic method can reduce computation cost and is validated through numerical simulations.
APPLIED MATHEMATICS AND COMPUTATION
(2021)
Article
Mathematics, Applied
Huahai Qiu, Li Wan, Zhigang Zhou, Qunjiao Zhang, Qinghua Zhou
Summary: This article investigates the global exponential periodicity of nonlinear neural networks with multiple time-varying delays. Due to the presence of multiple delays, such neural networks cannot be written in the vector-matrix form. Although the neural network with multiple time-varying delays has been studied using the Lyapunov-Krasovskii functional method in the literature, sufficient conditions in the linear matrix inequality form have not been obtained. Two sets of sufficient conditions in the linear matrix inequality form are established to ensure that two arbitrary solutions of the neural network with multiple delays attract each other exponentially. This is a key prerequisite to prove the existence, uniqueness, and global exponential stability of periodic solutions. Examples are provided to demonstrate the effectiveness of the established results, and a comparison with previous results shows their inapplicability to the systems in these examples.
Article
Automation & Control Systems
Bing Li, Yuwei Cao, Yongkun Li
Summary: In this paper, a class of octonion-valued stochastic recurrent neural networks with time-varying delays is considered. The existence, uniqueness, and global exponential stability of almost automorphic solutions in distribution are proved using Banach fixed point theorem and inequality technique. The results obtained in this study are new. An illustrative example is also provided to demonstrate the effectiveness of the results.
INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Xi-Le Zhao, Jing-Hua Yang, Tian-Hui Ma, Tai-Xiang Jiang, Michael K. Ng, Ting-Zhu Huang
Summary: This work proposes a novel tensor completion framework that can simultaneously utilize global-local-nonlocal priors by adopting the tensor train rank to characterize global correlation and incorporating Plug-and-Play denoisers to preserve local details and exploit nonlocal self-similarity. A proximal alternating minimization algorithm is designed to efficiently solve the model under the Plug-and-Play framework. Extensive experiments demonstrate the organic benefits of these priors, achieving state-of-the-art performance quantitatively and qualitatively.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Mathematics, Applied
Ben-Zheng Li, Xi-Le Zhao, Teng-Yu Ji, Xiong-Jun Zhang, Ting-Zhu Huang
Summary: The proposed method combines nonlinear and linear semi-orthogonal transforms to achieve low-rank tensor completion, introduces a new low-rankness metric and optimization model, and designs a corresponding algorithm. Experimental results demonstrate that the method outperforms the current state-of-the-art linear transform methods in image and video processing in terms of performance metrics and visual quality.
JOURNAL OF SCIENTIFIC COMPUTING
(2022)
Article
Geochemistry & Geophysics
Hao Zhang, Ting-Zhu Huang, Xi-Le Zhao, Wei He, Jae Kyu Choi, Yu-Bang Zheng
Summary: In this article, we propose a transformed domain model called T-RSTR, which combines sparse and low-tensor-ring (TR)-rank priors for hyperspectral image denoising. T-RSTR integrates the transform-based low-TR-rank and sparse regularizers to capture the global low-rankness and sparsity of the transformed tensors, resulting in significantly improved denoising performance. An elaborately designed proximal alternating minimization-based algorithm is used to solve the T-RSTR model, and its convergence is theoretically proved. Extensive numerical results demonstrate the superiority of T-RSTR over competing methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Jin-Fan Hu, Ting-Zhu Huang, Liang-Jian Deng, Tai-Xiang Jiang, Gemine Vivone, Jocelyn Chanussot
Summary: The paper introduces a deep convolutional neural network architecture to fuse low-resolution HSI and high-resolution multispectral image for generating high-resolution HSI. By preserving spatial and spectral information using LR-HSI and HR-MSI, and utilizing attention and pixelShuffle modules for high-quality spatial details extraction, the proposed network achieves the best performance compared to recent HSI super-resolution approaches.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Automation & Control Systems
Hao Zhang, Xi-Le Zhao, Tai-Xiang Jiang, Michael K. Ng, Ting-Zhu Huang
Summary: This study proposes a novel method of utilizing multiscale feature tensorization to recover missing values of multidimensional images at the feature level, and leveraging the low-rankness of the resulting tensorization to benefit subsequent image applications.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Artificial Intelligence
Tian-Jiang Zhang, Liang-Jian Deng, Ting-Zhu Huang, Jocelyn Chanussot, Gemine Vivone
Summary: In this article, a novel deep neural network architecture called TDNet is proposed for pansharpening, which utilizes the spatial details of the panchromatic image to enhance the spatial resolution of the multispectral image. Experimental results demonstrate the superiority of TDNet compared to other methods.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Geochemistry & Geophysics
Jian-Li Wang, Xi-Le Zhao, Heng-Chao Li, Ke-Xiang Cao, Jiaqing Miao, Ting-Zhu Huang
Summary: Cloud removal is crucial for downstream applications in remotely sensed images (RSIs) processing. Multitemporal RSIs (MRSIs), with abundant spatial-spectral-temporal (SST) information, offer new opportunities for cloud removal. We propose an Unsupervised Domain Factorization Network (UnDFN) that effectively and efficiently exploits the rich SST information of MRSIs by leveraging the low-rankness.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Li-Yuan Li, Ting-Zhu Huang, Yu-Bang Zheng, Wen-Jie Zheng, Jie Lin, Guo-Cheng Wu, Xi-Le Zhao
Summary: In this study, a method based on gradient domain fidelity and guided gradient is proposed to remove thick clouds in multitemporal remote sensing images. The guided gradient of the cloudy region is estimated using the Regression method from the cloud-free region of different time nodes. An efficient proximal alternating minimization (PAM)-based algorithm is developed to solve the proposed nonconvex model. Extensive simulated and real experiments demonstrate that the proposed method outperforms its competitors in preserving fine edges and textures.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Zhong-Cheng Wu, Ting-Zhu Huang, Liang-Jian Deng, Jie Huang, Jocelyn Chanussot, Gemine Vivone
Summary: In this paper, a novel method for multispectral image pansharpening, called LRTCFPan, is proposed based on low-rank tensor completion (LRTC) with some regularizers. By introducing the LRTC technique and utilizing the ISR model, the pansharpening problem is effectively solved. Furthermore, the low-tubal-rank prior of multispectral images is investigated to improve the completion and global characterization. Experimental results demonstrate that the LRTCFPan method outperforms other state-of-the-art pansharpening methods on reduced-resolution and full-resolution data.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Engineering, Electrical & Electronic
Wei-Hao Wu, Ting-Zhu Huang, Hao Zhang, Jian-Li Wang, Xi-Le Zhao
Summary: This paper proposes an untrained low-rank neural network prior (ULRNNP) for multi-dimensional image recovery, which has powerful representation ability and stable behavior. By using a nonlinear Tucker decomposition module, ULRNNP can design a friendly stopping criteria without the need for a reference ground truth image.
IEEE SIGNAL PROCESSING LETTERS
(2023)
Article
Geochemistry & Geophysics
Jin-Liang Xiao, Ting-Zhu Huang, Liang-Jian Deng, Zhong-Cheng Wu, Xiao Wu, Gemine Vivone
Summary: Pansharpening is a method to fuse multispectral and panchromatic images to generate a high-resolution multispectral image. Previous pixel-based methods suffer from spatial distortions, while the proposed SFNLR method exploits the local smoothness and nonlocal self-similarity of the coefficients to improve accuracy.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Ting Xu, Ting-Zhu Huang, Liang-Jian Deng, Naoto Yokoya
Summary: In this article, an iterative regularization method based on tensor subspace representation (IR-TenSR) is proposed for hyperspectral image super-resolution (HSI-SR). A tensor subspace representation (TenSR)-based regularization model is introduced, which integrates the global spectral-spatial low-rank and the nonlocal self-similarity priors of high spatial resolution hyperspectral image (HR-HSI). An iterative regularization procedure is designed to utilize the residual information of low-resolution images, and an effective algorithm based on the proximal alternating minimization method is developed to solve the TenSR-regularization model. Experimental results demonstrate the superiority of IR-TenSR compared to state-of-the-art fusion approaches.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Jian-Li Wang, Ting-Zhu Huang, Xi-Le Zhao, Yu-Chun Miao
Summary: This paper presents a video completion model that addresses the challenge of effectively capturing implicit low-rankness in complex video completion. By combining deterministic low-rank prior and deep image prior, the proposed model outperforms some state-of-the-art tensor completion methods in complex video completion, as verified by experimental results.
IEEE SIGNAL PROCESSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Wei-Hao Wu, Ting-Zhu Huang, Xi-Le Zhao, Jian-Li Wang, Yu-Bang Zheng
Summary: This study proposes a novel model for hyperspectral image (HSI) denoising, combining the tensor low-rank prior and the deep spatial-spectral prior to capture both the global structure and local details of the underlying HSI. Experimental results demonstrate the advantages of the proposed model in preserving local details and the global structure of the HSI.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Ping Yang, Ting-Zhu Huang, Jie Huang, Jin-Ju Wang
Summary: The proposed method incorporates weighted nuclear norm and $L_{1/2}$ norm to consider the low-rank and sparse priors of each abundance map simultaneously in hyperspectral unmixing. An adaptive update mechanism is implemented to treat each constraint differently, leading to improved unmixing effect and solving speed.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.