A Decentralized Primal-Dual Framework for Non-Convex Smooth Consensus Optimization

Gabriel Mancino-Ball,Yangyang Xu,Jie Chen

IEEE Transactions on Signal Processing（2023）

引用 0|浏览25

暂无评分

摘要

In this work, we introduce ADAPD, A D ecentr A lized P rimal- D ual algorithmic framework for solving non-convex and smooth consensus optimization problems over a network of distributed agents. The proposed framework relies on a novel problem formulation that elicits ADMM-type updates, where each agent first inexactly solves a local strongly convex subproblem with any method of its choice and then performs a neighbor communication to update a set of dual variables. We present two variants that allow for a single gradient step for the primal updates or multiple communications for the dual updates, to exploit the tradeoff between the per-iteration cost and the number of iterations. When multiple communications are performed, ADAPD can achieve theoretically optimal communication complexity results for non-convex and smooth consensus problems. Numerical experiments on several applications, including a deep-learning one, demonstrate the superiority of ADAPD over several popularly used decentralized methods.

查看译文

关键词

Non-convex consensus optimization,decentralized optimization,primal-dual method,decentralized learning

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要