9期

最新文章

• Convergence of Energy-Based Learning in Linear Resistive Networks AI翻译
  作者：Huijzer, Anne-Men; Chaffey, Thomas; Besselink, Bart; van Waarde, Henk J.
  作者单位：University of Groningen; University of Sydney
  摘要：Energy-based learning algorithms are alternatives to backpropagation and are well-suited to distributed implementations in analog electronic devices. However, a rigorous theory of convergence is lacking. We make a first step in this direction by analyzing a particular energy-based learning algorithm, contrastive learning, applied to a network of linear adjustable resistors. It is shown that, in this setup, contrastive learning is equivalent to projected gradient descent on a convex function wi...
• Distributed Online Optimization Over Partially Free-In and Free-Out Networks AI翻译
  作者：Liu, Yuxuan; Ye, Maojiao; Ding, Lei; Han, Qing-Long
  作者单位：Nanjing University of Science & Technology; Nanjing University of Posts & Telecommunications; Nanjing University of Posts & Telecommunications; Swinburne University of Technology
  摘要：The article studies a distributed online optimization problem over partially free-in and free-out networks, in which a set of unfixed agents cooperate to minimize the sum of a group of time-varying functions over a time horizon. To be specific, the agents are divided into static agents and dynamic agents. The static agents are those who remain in the network during the whole time horizon, while the dynamic agents are allowed to join and leave the network freely. Based on the dual averaging tec...
• Distributed Resilient Fixed-Time Control for Cooperative Output Regulation of MASs Over Directed Graphs Under DoS Attacks AI翻译
  作者：Cao, Wenji; Liu, Lu; Zhang, Dan; Feng, Gang
  作者单位：City University of Hong Kong; Zhejiang University of Technology
  摘要：This article addresses the problem of fixed-time cooperative output regulation for linear multiagent systems over directed graphs under denial-of-service attacks. A novel distributed resilient fixed-time controller is developed that comprises a distributed resilient fixed-time observer taking general directed graphs into consideration and a distributed resilient fixed-time control law for each agent. The proposed controller neither depends on Laplacian symmetry nor requires strong connectivity...
• Dynamical System Approach for Optimal Control Problems With Equilibrium Constraints Using Gap-Constraint-Based Reformulation AI翻译
  作者：Lin, Kangyu; Ohtsuka, Toshiyuki
  作者单位：Kyoto University
  摘要：This study focuses on using direct methods (first-discretize-then-optimize) to solve optimal control problems for a class of nonsmooth dynamical systems governed by differential variational inequalities (DVI), called optimal control problems with equilibrium constraints (OCPECs). In the discretization step, we propose a class of novel approaches to smooth the DVI. The generated smoothing approximations of the DVI, referred to as gap-constraint-based reformulations, have computational advantage...
• Communication-Efficient Distributed Online Nonconvex Optimization With Time-Varying Constraints AI翻译
  作者：Zhang, Kunpeng; Xu, Lei; Yi, Xinlei; Wen, Guanghui; Cao, Ming; Johansson, Karl H.; Chai, Tianyou; Yang, Tao
  作者单位：Northeastern University - China; Royal Institute of Technology; Southeast University - China; University of Groningen
  摘要：This article considers distributed online nonconvex optimization with time-varying inequality constraints, where the nonconvex local loss and convex local constraint functions can vary arbitrarily across iterations. For a time-varying directed graph, we propose two distributed bandit online primal-dual algorithms with compressed communication to efficiently utilize communication resources in the one-point and two-point bandit feedback settings, respectively. To measure the performance of the p...
• Safety-Constrained Control With Tunable Performance Using Dual-Mode Method and Local Trajectory Adjustment AI翻译
  作者：Zhao, Xingqiang; Song, Yongduan; Wen, Changyun
  作者单位：Chongqing University; Lingnan University; Nanyang Technological University
  摘要：This article presents a novel safety-constrained control framework for a class of uncertain high-order multi-input-multi-output (MIMO) nonlinear systems, unifying trajectory correction strategies with tunable performance constraint designs. In contrast to existing methods that assume the desired trajectory is always safe and feasible, we introduce a localized trajectory correction mechanism based on a compact-support transition function. This mechanism achieves intended adjustments by superimp...
• Algorithm Design and Regret Analysis for Aggregative Optimization in Open Multiagent Systems AI翻译
  作者：Liu, Yuxuan; Ye, Maojiao; Ding, Lei; Han, Qing-Long
  作者单位：Nanjing University of Science & Technology; Nanjing University of Posts & Telecommunications; Nanjing University of Posts & Telecommunications; Swinburne University of Technology
  摘要：This article studies the problem of aggregative optimization in open multiagent systems (OMAS), where agents are allowed to join and leave the system in a free manner during the decision-making process. A multiaggregator communication mechanism is proposed to facilitate information exchange among agents, in which the aggregators are responsible for collecting information from agents and exchanging it with neighboring aggregators. Based on the multiaggregator communication mechanism, a novel se...
• Finite Time Convergence Analysis of Policy Gradient Learning in Stochastic Stackelberg Games AI翻译
  作者：Das, Pranoy; Zaman, Muhammad Aneeq uz; Gupta, Vijay
  作者单位：Purdue University System; Purdue University; University of Illinois System; University of Illinois Urbana-Champaign
  摘要：Designing and analyzing learning algorithms for general sum stochastic Stackelberg games remain challenging. We propose an inner-outer loop policy gradient-based learning algorithm for this problem and analyze its finite time convergence. Our analysis does not assume a time scale separation between the leader and the follower or a special structure of the game, such as being Markov potential or zero-sum.
• High-Perception-Area Collaborative Exploration of Double-Integrator Robots in Parabolic-Type Diffusion Fields With Partially Unknown Information AI翻译
  作者：Li, Rui-Guo; Wang, Liang; Wu, Huai-Ning
  作者单位：Tiangong University; Shenzhen University; Beihang University
  摘要：This article solves a high-perception-area collaborative exploration issue for double-integrator robots in 2-dimensional parabolic-type diffusion scalar fields with partially unknown information. In the absence of surrounding field density information for robots, a distributed field density observer with adaptive update laws and event-triggering mechanisms is built to approximate necessary field density information. Taking into account of the scenario that robot's actual velocity is unmeasurab...
• Online Learning and Control Synthesis for Reachable Paths of Unknown Nonlinear Systems AI翻译
  作者：Meng, Yiming; Shafa, Taha; Wei, Jesse; Ornik, Melkior
  作者单位：University of Illinois System; University of Illinois Urbana-Champaign; University of Illinois System; University of Illinois Urbana-Champaign; University of Illinois System; University of Illinois Urbana-Champaign; University of Illinois System; University of Illinois Urbana-Champaign
  摘要：In this article, we present a novel method to drive a nonlinear system to a desired state, with limited a priori knowledge of its dynamic model: local dynamics at a single point and the bounds on the rate of change of these dynamics. This method synthesizes control actions by utilizing locally learned dynamics along a trajectory, based on data available up to that moment, and known proxy dynamics, which can generate an underapproximation of the unknown system's true reachable set. An important...