2024

最新文章

• Adaptive Boundary Observers for Hyperbolic PDEs With Application to Traffic Flow Estimation AI翻译
  作者：Wu, Jiahao; Zhan, Jingyuan; Zhang, Liguo
  作者单位：Beijing University of Technology
  摘要：This article studies the problem of adaptive boundary observer design for a class of linear hyperbolic partial differential equations (PDEs) subject to in-domain and boundary parameter uncertainties. Based on the swapping transformation technique, a Luenberger-type boundary observer with the least squares parameter estimation law is designed, which relies only on the measurements at boundaries of the system. By employing the Lyapunov function method, we prove that the exponential convergence o...
• BIG Hype: Best Intervention in Games via Distributed Hypergradient Descent AI翻译
  作者：Grontas, Panagiotis D.; Belgioioso, Giuseppe; Cenedese, Carlo; Fochesato, Marta; Lygeros, John; Dorfler, Florian
  作者单位：Swiss Federal Institutes of Technology Domain; ETH Zurich
  摘要：Hierarchical decision making problems, such as bilevel programs and Stackelberg games, are attracting increasing interest in both the engineering and machine learning communities. Yet, existing solution methods lack either convergence guarantees or computational efficiency, due to the absence of smoothness and convexity. In this work, we bridge this gap by designing a first-order hypergradient-based algorithm for Stackelberg games and mathematically establishing its convergence using tools fro...
• A Decentralized Control Synthesis Approach for Networked Systems With Arbitrary Interconnections AI翻译
  作者：Welikala, Shirantha; Lin, Hai; Antsaklis, Panos J.
  作者单位：Stevens Institute of Technology; University of Notre Dame
  摘要：This article considers the problem of decentralized analysis and control synthesis to verify and enforce properties like stability and dissipativity of large-scale networked systems comprised of linear subsystems interconnected in an arbitrary topology. In particular, we design systematic networked system analysis and control synthesis processes that can be executed in a decentralized manner with minimal information sharing among the subsystems. We also show that, for such decentralized proces...
• A Family of Hyperparameter Estimators Linking EB and SURE for Kernel-Based Regularization Methods AI翻译
  作者：Zhang, Meng; Chen, Tianshi; Mu, Biqiang
  作者单位：Chinese Academy of Sciences; Academy of Mathematics & System Sciences, CAS; The Chinese University of Hong Kong, Shenzhen; The Chinese University of Hong Kong, Shenzhen; Shenzhen Research Institute of Big Data; The Chinese University of Hong Kong, Shenzhen
  摘要：Hyperparameter estimation is a critical aspect of kernel-based regularization methods (KRMs), alongside kernel design. Empirical Bayes (EB) and Stein's unbiased risk estimator (SURE) are two widely used hyperparameter estimators for tuning the unknown hyperparameters associated with the kernel matrix of KRMs. However, EB and SURE exhibit different characteristics in both theory and practice. Theoretically, SURE is asymptotically optimal in terms of minimizing the mean squared error (MSE), wher...
• Asynchronous Distributed Reinforcement Learning for LQR Control via Zeroth-Order Block Coordinate Descent AI翻译
  作者：Jing, Gangshan; Bai, He; George, Jemin; Chakrabortty, Aranya; Sharma, Piyush K.
  作者单位：Chongqing University; Oklahoma State University System; Oklahoma State University - Stillwater; North Carolina State University
  摘要：Recently introduced distributed zeroth-order optimization (ZOO) algorithms have shown their utility in distributed reinforcement learning (RL). Unfortunately, in the gradient estimation process, almost all of them require random samples with the same dimension as the global variable and/or require evaluation of the global cost function, which may induce high estimation variance for large-scale networks. In this paper, we propose a novel distributed zeroth-order algorithm by leveraging the netw...
• Duality of Ensemble Systems Through Moment Representations AI翻译
  作者：Narayanan, Vignesh; Zhang, Wei; Li, Jr-Shin
  作者单位：Washington University (WUSTL); University of South Carolina System; University of South Carolina Columbia
  摘要：Controlling large-scale dynamic population systems, known as ensemble control, is a pervasive and essential task in many emerging applications from diverse scientific domains. Previous focuses in the area of ensemble control have been placed on seeking open-loop control strategies due to the unavailability of state feedback information for each individual system in the ensemble. In this article, we develop a foundational framework for analysis and control of ensemble systems with closed feedba...
• Online Optimization and Ambiguity-Based Learning of Distributionally Uncertain Dynamic Systems AI翻译
  作者：Li, Dan; Fooladivanda, Dariush; Martinez, Sonia
  作者单位：University of California System; University of California San Diego; University of California System; University of California Berkeley
  摘要：This article proposes a novel approach to construct data-driven online solutions to optimization problems (P) subject to a class of distributionally uncertain dynamical systems. The introduced framework allows for the simultaneous learning of distributional system uncertainty via a parameterized, control-dependent ambiguity set using a finite historical dataset, and its use to make online decisions with probabilistic regret function bounds. Leveraging the merits of machine learning, the main t...
• A Preset-Trajectory-Based Singularity-Free Preassigned Performance Control Approach AI翻译
  作者：Shi, Wenrui; Hou, Mingzhe; Duan, Guangren
  作者单位：Harbin Institute of Technology
  摘要：The preassigned performance control (PPC) methods have attracted considerable attention in recent years; however, most of the mainstream PPC methods utilize barrier functions, and thus, may suffer from the singularity problem of the control law under some unexpected conditions, such as sensor faults. In this article, a new robust PPC method without using barrier functions is proposed for second-order vector nonlinear systems, which can completely avoid the singularity problem of the control la...
• Forgetting-Factor Regrets for Online Convex Optimization AI翻译
  作者：Liu, Yuhang; Zhao, Wenxiao; Yin, George
  作者单位：Chinese Academy of Sciences; Academy of Mathematics & System Sciences, CAS; Chinese Academy of Sciences; University of Chinese Academy of Sciences, CAS; University of Connecticut
  摘要：This article develops a class of novel algo-rithms for online convex optimization. The key constructis a forgetting-factor regret. It introduces weights to theobjective functions at each time instanttand allows theweights of the past objective functions decaying to zero.We establish the forgetting-factor regret bounds of clas-sical algorithms including online gradient descent algo-rithms, online gradient-free algorithms, and online Frank-Wolfe algorithms. In addition, the article introduces on...
• Optimal Control of Two-Dimensional Roesser Model: Solution Based on Reinforcement Learning AI翻译
  作者：Ye, Linwei; Zhao, Zhonggai; Liu, Fei
  作者单位：Jiangnan University; Jiangnan University
  摘要：This article addresses the infinite-region linear quadratic regulation problem of the discrete two-dimensional (2-D) Roesser model, drawing inspiration from reinforcement learning principles. It introduces a novel proof establishing that expressing the optimal control law in 2-D through state feedback is unattainable. Subsequently, a policy iteration framework is proposed to derive suboptimal state feedback, followed by an exploration of the true optimal policy through value iteration. The eff...