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
Automation & Control Systems
Qiwei Liu, Huaicheng Yan, Hao Zhang, Xisheng Zhan, Kaibo Shi
Summary: This article investigates the problem of exponential synchronization for memristor-based neural networks with mixed time-varying delays and parameter perturbations. A periodically intermittent control protocol is designed to guarantee the exponential synchronization between two networks. The exponential synchronization criteria for these networks under the proposed controller are obtained using nonsmooth analysis, Halanay inequality, and Lyapunov theory. The synchronization of these networks is considered within the framework of a second-order system directly, which is different from existing literature.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Automation & Control Systems
Meng Hui, Jiefei Yan, Jiao Zhang, Herbert Ho-Ching Iu, Rui Yao, Lin Bai
Summary: This article proposes a novel hybrid control scheme for achieving exponential synchronization of inertial neural networks with mixed delays, and verifies the effectiveness of the method through numerical simulations.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2022)
Article
Computer Science, Artificial Intelligence
Yao Xu, Fu Sun, Wenxue Li
Summary: This paper investigates the issue of exponential synchronization of fractional-order multilayer coupled neural networks with reaction-diffusion terms using periodically intermittent control. Theoretical results show that the exponential convergence rate depends on the control gain and the order of fractional derivative. An illustrative numerical example is provided to further verify the feasibility and effectiveness of the results.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Mathematics, Applied
Beibei Guo, Yu Xiao
Summary: This paper studies exponential synchronization for hybrid multi-weighted complex networks by considering both Markov jump and reaction-diffusion effects. Several synchronization criteria are derived using network split technique, graph theory, and Lyapunov method, showing the effects of multiple weights, Markov jump, and reaction-diffusion. An application to Cohen-Grossberg neural networks is also conducted with corresponding synchronization criterion. Numerical simulations are presented to demonstrate the effectiveness of the obtained theoretical results.
MATHEMATICAL METHODS IN THE APPLIED SCIENCES
(2023)
Article
Automation & Control Systems
Jianwen Feng, Jingxin Chen, Jingyi Wang, Yi Zhao
Summary: This paper investigates the exponential synchronization problem for complex networks with hybrid delays using the event intermittent control (EIC) strategy. A modified lemma related to delays is derived without pre-designing intermittent instants, and synchronization criteria with less conservatism are established using linear matrix inequalities (LMIs). The effectiveness of the proposed EIC strategy is verified through simulations of a numerical example.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2022)
Article
Mathematics, Applied
Chenguang Xu, Minghui Jiang, Junhao Hu
Summary: This paper focuses on the mean-square finite-time synchronization (MFTS) problem of stochastic competitive neural networks with infinite time-varying discrete delays and reaction-diffusion terms (IRSCNNs). A new approach, which uses integral inequality, Gronwall-type inequality, and comparison strategy, is proposed to study MFTS. Two control schemes, a feedback control scheme and a new adaptive control strategy, are designed to achieve MFTS of IRSCNNs. The correctness of the results is verified through examples.
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION
(2023)
Article
Computer Science, Information Systems
Yinping Xie, Ling Xiao, Leimin Wang, Gaohua Wang
Summary: This paper addresses the stability problem of genetic regulatory networks with reaction diffusion by deriving global exponential stability criteria in algebraic form for networks with discrete and distributed delays. The stability conditions are simple, universal, and can be directly calculated using network parameters. Numerical simulations are used to illustrate the validity and feasibility of the results.
Article
Computer Science, Artificial Intelligence
Zhenyuan Guo, Shiqin Wang, Jun Wang
Summary: This article presents new theoretical results on global exponential synchronization of nonlinear coupled delayed memristive neural networks with reaction-diffusion terms and Dirichlet boundary conditions. Two control schemes are introduced to ensure the synchronization, and sufficient criteria are derived through the utilization of Lyapunov stability theorem and Divergence theorem. Two illustrative examples are provided to substantiate the theoretical results and showcase the advantages and disadvantages of the control schemes.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Jiejie Chen, Boshan Chen, Zhigang Zeng
Summary: This paper introduces a new intermittent event-triggered control method (IETC), and discusses the quasi-synchronization problem of coupled memristive neural networks with time-varying delays, establishing criteria for both synchronization and quasi-synchronization.
Article
Computer Science, Artificial Intelligence
Qian Cui, Lulu Li, Wei Huang
Summary: This paper mainly focuses on the synchronization of coupled delayed reaction-diffusion neural networks with delayed impulses. The direct error method is used to study the synchronization, and a strict comparison principle is derived to solve the synchronization of systems with delayed impulses. Combined with the average impulsive delay method, some sufficient conditions for synchronization are proposed, indicating the positive role of impulsive delay. Two examples are provided to illustrate the validity of the results.
NEURAL PROCESSING LETTERS
(2023)
Article
Automation & Control Systems
Ning Zhang, Shunjie Huang, Wenxue Li
Summary: This paper investigates the pth moment exponential stability of stochastic delayed systems, taking into account both semi-Markov jumps and stochastic mixed impulses. It establishes new impulsive differential inequalities with semi-Markov jumps and stochastic mixed impulses. By cleverly combining graph theory, stochastic analysis techniques, and the Lyapunov method, stability criteria for stochastic delayed semi-Markov jump systems with stochastic mixed impulses are proposed. Finally, the theoretical results are applied to oscillator systems, and the simulation results confirm the effectiveness of the theoretical findings.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Xing-Yu Li, Qing-Ling Fan, Xiao-Zhen Liu, Kai-Ning Wu
Summary: This article investigates the exponential stability of delay reaction-diffusion cellular neural networks (DRDCNNs) in two cases: when the state information is fully available and when it is not fully available. Aperiodically intermittent boundary controllers are designed to stabilize the controlled system when the state information is fully available, and observer-based aperiodically intermittent boundary controllers are proposed when the state information is not fully available. By utilizing the Lyapunov functional method and Poincare's inequality, a criterion for achieving exponential stabilization of DRDCNNs is obtained. The influence of diffusion coefficient matrix, control gains, time-delays, and control proportion on stability is studied based on the obtained results. Numerical examples are presented to illustrate the effectiveness of the theoretical results.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Qiwei Liu, Huaicheng Yan, Hao Zhang, Chaoyang Chen, Yufang Chang
Summary: This article addresses the exponential synchronization problem for a class of fuzzy inertial memsirtor-based neural networks with mixed time-varying delays. First, the inertial items are described as second-order systems and transformed into first-order systems by utilizing a appropriate variable substitution. Then, the fuzzy state-feedback control strategy and fuzzy adaptive control strategy are designed to ensure the exponential synchronization under the framework of Filippov solutions. The exponential synchronization algebraic conditions are obtained by choosing a proper Lyapunov-Krasovskii functional. Finally, two numerical simulations are provided to validate the effectiveness and benefit of the proposed results.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Xiaofan Li, Nikhil R. Pal, Huiyuan Li, Tingwen Huang
Summary: This article discusses and resolves the issue of intermittent event-triggered exponential stabilization for state-dependent switched fuzzy neural networks with mixed delays. An intermittent event-triggered control strategy is proposed to reduce the amount of samplings and save control costs. The theoretical analysis results are verified through a simulation example.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Mathematics, Applied
Juanping Yang, Hong-Li Li, Long Zhang, Cheng Hu, Haijun Jiang
Summary: This paper investigates the problems of quasi-projective synchronization (QPS) and finite-time synchronization (FTS) for a type of delayed fractional-order BAM neural networks (DFOBAMNNs). By designing fresh quantized controllers, the goals of synchronization are achieved and the settling time and error level are accurately evaluated.
MATHEMATICAL METHODS IN THE APPLIED SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Juanping Yang, Hong-Li Li, Long Zhang, Cheng Hu, Haijun Jiang
Summary: This paper investigates synchronization analysis and parameters identification for uncertain delayed fractional-order BAM neural networks. By designing control strategies and parameters updated laws, using Lyapunov function theory, fractional calculus theory and inequality analysis techniques, the paper establishes criteria for ensuring finite-time synchronization and Mittag-Leffler synchronization of the considered networks. The settling time of finite-time synchronization is also provided, which is related to the initial values. Furthermore, parameter identification is successfully realized for uncertain or unknown parameters. Numerical examples are provided to demonstrate the effectiveness of the theoretical results.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Automation & Control Systems
Tingting Zhao, Cheng Hu, Juan Yu, Leimin Wang, Haijun Jiang
Summary: This paper focuses on the fixed/preassigned-time (FXT/PAT) synchronization of multilayered networks with heterogeneous self-dynamics of nodes and arbitrary prescribed smooth orbit for synchronized state. The augmented network allows for the removal of topological connectivity limitations and reduction of conservatism in synchronization conditions. Continuous control protocols are developed to achieve FXT synchronization, with effective criteria established using the theorem of FXT stability. Additionally, the relationship between synchronized time estimation and layer parameter is revealed. Moreover, the PAT synchronization is investigated with two control schemes and finite control gains for a preassigned synchronized time. The developed control designs and criteria are validated through numerical simulations.
Article
Computer Science, Artificial Intelligence
Xuejiao Qin, Haijun Jiang, Jianlong Qiu, Cheng Hu, Yue Ren
Summary: This article investigates the fixed-time (F-T) and predefined-time (P-T) cluster lag synchronization of stochastic multi-weighted complex networks (SMWCNs) using strictly intermittent quantized control (SIQC). A novel F-T stability lemma is proved, and an accurate estimation of settling time (ST) is obtained. Simple conditions for F-T cluster lag synchronization are derived using the proposed F-T stability and a SIQC strategy. The article also explores P-T cluster lag synchronization based on a SIQC design, with the settling time predefined by an adjustable constant.
Article
Mathematics, Applied
Zhao Jiang, Ahmadjan Muhammadhaji, Cheng Hu, Zhidong Teng
Summary: This paper investigates N-species Lotka-Volterra cooperation models with feedback controls and continuous delays. The global attractiveness, permanence, and periodic solutions of the model are analyzed using integral inequality techniques, a comparison principle, and the Lyapunov-Razumikhin method. Additionally, a numerical example is conducted to validate the feasibility and practicality of the proposed results.
QUALITATIVE THEORY OF DYNAMICAL SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Duo Zhang, Deqiang Ouyang, Lan Shu, Cheng Hu, Kaibo Shi, Shiping Wen
Summary: This paper proposes an event-based dynamic output feedback control method to achieve fuzzy neural network synchronization under mixed delay and deception attacks. A dynamic event-triggered mechanism is developed to save communication resources while ensuring system performance. The effectiveness of the method is demonstrated through numerical tests.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Zhanheng Chen, Kailong Xiong, Cheng Hu
Summary: This paper addresses the fixed/preassigned-time synchronization problem for a type of full complex-variable BAM neural networks (BAMNNs), using direct analysis method instead of the classical decomposition approach. The paper proposes effective complex-valued control strategies based on the complex-valued signum function and two different forms of norms, which ensure fixed-time synchronization with less conservatism compared to previous results. The paper also considers preassigned-time synchronization by designing complex-valued control laws with bounded control gains, allowing the synchronization time to be arbitrarily preset within the allowable range. Numerical examples are provided to support the theoretical results.
NEURAL PROCESSING LETTERS
(2023)
Article
Computer Science, Information Systems
Liang Feng, Cheng Hu, Quanxin Zhu, Fanchao Kong, Shiping Wen
Summary: This article focuses on the fixed/preassigned-time (FIX/PT) output synchronization of homogeneous complex networks with output coupling. By introducing observable output coupling and distributed dynamic event-triggered control schemes, it reduces communication frequency and resource waste, and provides concise criteria for achieving synchronization.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Ziwei Guo, Jinshui Ren, Zhen Liu, Xuzheng Liu, Cheng Hu
Summary: This article mainly focuses on fixed-time (FXT) synchronization of fully quaternion-valued memristive neural networks (QVMNNs) with generalized delays. The discontinuity issue of quaternion-valued memristive connection weights is effectively solved using set-valued mapping theory and the nonsmooth approach. A direct non-decomposing method is proposed by introducing the signum function, absolute-like norm, and quadratic norm in the quaternion field. Exponential-type controllers with simplified forms are designed, and criteria for FXT synchronization and the upper bound of the setting time are derived. Numerical examples confirm the achieved synchronization results.
Article
Automation & Control Systems
Liang Feng, Cheng Hu, Juan Yu, Haijun Jiang
Summary: This paper mainly focuses on the pinning synchronization problem of directed networks under disconnected switching topology. The authors introduce the concept of uniformly directed spanning tree on average (UDSTA) and design a switching pinning controller to investigate the asymptotic synchronization of complex networks with a large number of switching topologies. They also propose a class of almost-period (AP) switching modes and derive the pinning synchronization criteria for complex networks under AP switching topology. Finally, numerical simulations of coupled neural networks are conducted to illustrate the theoretical results. (c) 2023 Elsevier Ltd. All rights reserved.
NONLINEAR ANALYSIS-HYBRID SYSTEMS
(2023)
Article
Automation & Control Systems
Fanchao Kong, Quanxin Zhu, Cheng Hu, Tingwen Huang
Summary: This article proposes fixed-time stability lemmas for the Filippov system using new inequality approaches. The method used does not require integration of the Lyapunov function $V$ on the two integral intervals, which is different from existing methods. New estimations of settling times are provided, and improvements are made to the steepness exponents of the Lyapunov function $V$. In order to study nondifferentiable delayed neural networks modeled by the Filippov system, a class of discontinuous uncertain Cohen-Grossberg neural networks (CGNNs) with mixed delays is formulated, and the existence of periodic solutions is proved before considering stability.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Engineering, Multidisciplinary
Fanchao Kong, Cheng Hu, Leimin Wang, Tingwen Huang
Summary: This paper examines the periodicity and fixed-time stabilization of discontinuous neural networks with time-varying leakage and discrete delays, which is a more general case. Firstly, the study on leakage-delay-dependent periodic solutions is achieved through set-valued mapping, coincidence theorem, Holder inequality, and supremum-infimum principle. Secondly, new fixed-time stability lemmas of Filippov systems are established by using integral inequality, which have relaxed assumptions and do not require additional parameters. Thirdly, leakage-delay-dependent results on fixed-time stabilization of the proposed neural networks are obtained under modified non-chattering control laws based on the new stability lemmas. The validity of the proposed results is demonstrated through numerical examples.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
Article
Mathematics, Interdisciplinary Applications
Yunzhan Bai, Juan Yu, Cheng Hu
Summary: This paper investigates the synchronization of fractional-order output-coupling multiplex networks (FOOCMNs). A type of fractional-order multiplex network is introduced, where the intra-layer and inter-layer couplings are described separately, and nodes communicate via their outputs. Sufficient conditions for achieving asymptotic synchronization are provided based on designed adaptive control, and a quantized adaptive controller is developed to improve the effective utilization rate of network resources.
FRACTAL AND FRACTIONAL
(2023)
Article
Automation & Control Systems
Fanchao Kong, Quanxin Zhu, Cheng Hu, Tingwen Huang
Summary: This article investigates the fixed-time synchronization of different dimensional Filippov systems and proposes new FxT stability lemmas. It reveals that the relationships between the control gains can lead to different settling times. Additionally, algebraic inequality conditions are provided to guarantee the FxT synchronization.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Mathematics
Shichao Jia, Cheng Hu, Haijun Jiang
Summary: This article investigates fixed-time synchronization and preassigned-time synchronization of Cohen-Grossberg quaternion-valued neural networks with discontinuous activation functions and generalized time-varying delays. It introduces a dynamic model in the quaternion field and designs two types of discontinuous controllers utilizing the quaternion-valued signum function. By developing a direct analytical approach and using the theory of non-smooth analysis, it derives several criteria for achieving fixed-time synchronization and estimates more precise convergence times. The article also addresses preassigned-time synchronization for practical requirements.
Article
Computer Science, Artificial Intelligence
Hamdan Abdellatef, Lina J. Karam
Summary: This paper proposes performing the learning and inference processes in the compressed domain to reduce computational complexity and improve speed of neural networks. Experimental results show that modified ResNet-50 in the compressed domain is 70% faster than traditional spatial-based ResNet-50 while maintaining similar accuracy. Additionally, a preprocessing step with partial encoding is suggested to improve resilience to distortions caused by low-quality encoded images. Training a network with highly compressed data can achieve good classification accuracy with significantly reduced storage requirements.
Article
Computer Science, Artificial Intelligence
Victor R. Barradas, Yasuharu Koike, Nicolas Schweighofer
Summary: Inverse models are essential for human motor learning as they map desired actions to motor commands. The shape of the error surface and the distribution of targets in a task play a crucial role in determining the speed of learning.
Article
Computer Science, Artificial Intelligence
Ting Zhou, Hanshu Yan, Jingfeng Zhang, Lei Liu, Bo Han
Summary: We propose a defense strategy that reduces the success rate of data poisoning attacks in downstream tasks by pre-training a robust foundation model.
Article
Computer Science, Artificial Intelligence
Hao Sun, Li Shen, Qihuang Zhong, Liang Ding, Shixiang Chen, Jingwei Sun, Jing Li, Guangzhong Sun, Dacheng Tao
Summary: In this paper, the convergence rate of AdaSAM in the stochastic non-convex setting is analyzed. Theoretical proof shows that AdaSAM has a linear speedup property and decouples the stochastic gradient steps with the adaptive learning rate and perturbed gradient. Experimental results demonstrate that AdaSAM outperforms other optimizers in terms of performance.
Article
Computer Science, Artificial Intelligence
Juntong Yun, Du Jiang, Li Huang, Bo Tao, Shangchun Liao, Ying Liu, Xin Liu, Gongfa Li, Disi Chen, Baojia Chen
Summary: In this study, a dual manipulator grasping detection model based on the Markov decision process is proposed. By parameterizing the grasping detection model of dual manipulators using a cross entropy convolutional neural network and a full convolutional neural network, stable grasping of complex multiple objects is achieved. Robot grasping experiments were conducted to verify the feasibility and superiority of this method.
Article
Computer Science, Artificial Intelligence
Miaohui Zhang, Kaifang Li, Jianxin Ma, Xile Wang
Summary: This paper proposes an unsupervised person re-identification (Re-ID) method that uses two asymmetric networks to generate pseudo-labels for each other by clustering and updates and optimizes the pseudo-labels through alternate training. It also designs similarity compensation and similarity suppression based on the camera ID of pedestrian images to optimize the similarity measure. Extensive experiments show that the proposed method achieves superior performance compared to state-of-the-art unsupervised person re-identification methods.
Article
Computer Science, Artificial Intelligence
Florian Bacho, Dominique Chu
Summary: This paper proposes a new approach called the Forward Direct Feedback Alignment algorithm for supervised learning in deep neural networks. By combining activity-perturbed forward gradients, direct feedback alignment, and momentum, this method achieves better performance and convergence speed compared to other local alternatives to backpropagation.
Article
Computer Science, Artificial Intelligence
Xiaojian Ding, Yi Li, Shilin Chen
Summary: This research paper addresses the limitations of recursive feature elimination (RFE) and its variants in high-dimensional feature selection tasks. The proposed algorithms, which introduce a novel feature ranking criterion and an optimal feature subset evaluation algorithm, outperform current state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Naoko Koide-Majima, Shinji Nishimoto, Kei Majima
Summary: Visual images observed by humans can be reconstructed from brain activity, and the visualization of arbitrary natural images from mental imagery has been achieved through an improved method. This study provides a unique tool for directly investigating the subjective contents of the brain.
Article
Computer Science, Artificial Intelligence
Huanjie Tao, Qianyue Duan
Summary: In this paper, a hierarchical attention network with progressive feature fusion is proposed for facial expression recognition (FER), addressing the challenges posed by pose variation, occlusions, and illumination variation. The model achieves enhanced performance by aggregating diverse features and progressively enhancing discriminative features.
Article
Computer Science, Artificial Intelligence
Zhenyi Wang, Pengfei Yang, Linwei Hu, Bowen Zhang, Chengmin Lin, Wenkai Lv, Quan Wang
Summary: In the face of the complex landscape of deep learning, we propose a novel subgraph-level performance prediction method called SLAPP, which combines graph and operator features through an innovative graph neural network called EAGAT, providing accurate performance predictions. In addition, we introduce a mixed loss design with dynamic weight adjustment to improve predictive accuracy.
Article
Computer Science, Artificial Intelligence
Yiyang Yin, Shuangling Luo, Jun Zhou, Liang Kang, Calvin Yu-Chian Chen
Summary: Medical image segmentation is crucial for modern healthcare systems, especially in reducing surgical risks and planning treatments. Transanal total mesorectal excision (TaTME) has become an important method for treating colon and rectum cancers. Real-time instance segmentation during TaTME surgeries can assist surgeons in minimizing risks. However, the dynamic variations in TaTME images pose challenges for accurate instance segmentation.
Article
Computer Science, Artificial Intelligence
Teng Cheng, Lei Sun, Junning Zhang, Jinling Wang, Zhanyang Wei
Summary: This study proposes a scheme that combines the start-stop point signal features for wideband multi-signal detection, called Fast Spectrum-Size Self-Training network (FSSNet). By utilizing start-stop points to build the signal model, this method successfully solves the difficulty of existing deep learning methods in detecting discontinuous signals and achieves satisfactory detection speed.
Article
Computer Science, Artificial Intelligence
Wenming Wu, Xiaoke Ma, Quan Wang, Maoguo Gong, Quanxue Gao
Summary: The layer-specific modules in multi-layer networks are critical for understanding the structure and function of the system. However, existing methods fail to accurately characterize and balance the connectivity and specificity of these modules. To address this issue, a joint learning graph clustering algorithm (DRDF) is proposed, which learns the deep representation and discriminative features of the multi-layer network, and balances the connectivity and specificity of the layer-specific modules through joint learning.
Article
Computer Science, Artificial Intelligence
Guanghui Yue, Guibin Zhuo, Weiqing Yan, Tianwei Zhou, Chang Tang, Peng Yang, Tianfu Wang
Summary: This paper proposes a novel boundary uncertainty aware network (BUNet) for precise and robust colorectal polyp segmentation. BUNet utilizes a pyramid vision transformer encoder to learn multi-scale features and incorporates a boundary exploration module (BEM) and a boundary uncertainty aware module (BUM) to handle boundary areas. Experimental results demonstrate that BUNet outperforms other methods in terms of performance and generalization ability.