返回
Distributed Mirror Descent for Nonconvex Constrained Optimization
DOI:10.1007/978-981-95-1103-7_29.png)
摘要
En 中文
本文关注一类具有以下关键特征的分布式在线约束优化问题:i) 多个耦合约束的复杂耦合特性;ii) 时变有向图的时间动态不平衡性;iii) 局部成本函数的非凸性。为有效应对这些复杂挑战,提出了一种原始-对偶近端镜像下降(PDPMD)算法。此外,采用辅助变量来抵消时变有向图引起的不平衡。进一步证明,在满足某些温和条件下,所提出的方法以次线性收敛率达到平稳点。最后,通过数值示例验证了所提出算法的有效性。
Keyword:
Distributed nonconvex optimization
Time-varying unbalanced digraphs
Multiple coupled constraints
Online learning
期刊
I
IF:
0
论文数:
31
被引数:
0
机构
引用论文
Distributed Global Optimization for a Class of Nonconvex Optimization With Coupled Constraints一类耦合约束非凸优化的分布式全局优化
Distributed Constrained Optimization Over Unbalanced Time-Varying Digraphs: A Randomized Constraint Solving Algorithm不平衡时变有向图上的分布式约束优化: 一种随机约束求解算法
Online distributed nonconvex optimization with stochastic objective functions: High probability bound analysis of dynamic regrets☆
AUTOMATICA
IF5.9
Linear Convergence of First- and Zeroth-Order Primal-Dual Algorithms for Distributed Nonconvex Optimization分布式非凸优化的一阶和零阶原始对偶算法的线性收敛性
Distributed online constrained nonconvex optimization in dynamic environments over directed graphs有向图上动态环境中的分布式在线约束非凸优化
SIGNAL PROCESSING
IF3.6
Online Distributed Optimization With Nonconvex Objective Functions Under Unbalanced Digraphs: Dynamic Regret Analysis不平衡有向图下非凸目标函数的在线分布式优化: 动态后悔分析

