Distributed Optimization With Global Constraints Using Noisy Measurements

• 成果类型: Article
• 署名作者: Mai, Van Sy; La, Richard J.; Zhang, Tao; Battou, Abdella
• 署名单位: National Institute of Standards & Technology (NIST) - USA; University System of Maryland; University of Maryland College Park
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2023.3277312
• 发表日期: 2024
• 页码: 1089-1096
• 关键词: Distributed optimization; penalty method; stochastic optimization

摘要

We propose a new distributed optimization algorithm for solving a class of constrained optimization problems in which the objective function is separable (i.e., the sum of local objective functions of agents), the optimization variables of distributed agents, which are subject to nontrivial local constraints, are coupled by global constraints, and only noisy observations are available to estimate (the gradients of) local objective functions. In many practical scenarios, agents may not be willing to share their optimization variables with others. For this reason, we propose a distributed algorithm that does not require the agents to share their optimization variables with each other; instead, each agent maintains a local estimate of the global constraint functions and shares the estimate only with its neighbors. These local estimates of constraint functions are updated using a consensus-type algorithm, whereas the local optimization variables of each agent are updated using a first-order method based on noisy estimates of gradient. We prove that, when the agents adopt the proposed algorithm, their optimization variables converge with probability 1 to an optimal point of an approximated problem based on the penalty method.