Distributed Adaptive Gradient Algorithm With Gradient Tracking for Stochastic Nonconvex Optimization

• 成果类型: Article
• 署名作者: Han, Dongyu; Liu, Kun; Lin, Yeming; Xia, Yuanqing
• 署名单位: Beijing Institute of Technology
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2024.3380710
• 发表日期: 2024
• 页码: 6333-6340
• 关键词: Cost function; vectors; Upper bound; Radio frequency; Convex functions; Sparse matrices; robots; Adaptive gradient algorithm; Distributed nonconvex optimization; gradient tracking (GT); Stochastic Gradient

摘要

This article considers a distributed stochastic nonconvex optimization problem, where the nodes in a network cooperatively minimize a sum of $L$-smooth local cost functions with sparse gradients. By adaptively adjusting the stepsizes according to the historical (possibly sparse) gradients, a distributed adaptive gradient algorithm is proposed, in which a gradient tracking estimator is used to handle the heterogeneity between different local cost functions. We establish an upper bound on the optimality gap, which indicates that our proposed algorithm can reach a first-order stationary solution dependent on the upper bound on the variance of the stochastic gradients. Finally, numerical examples are presented to illustrate the effectiveness of the algorithm.