A Unified Framework for Understanding Distributed Optimization Algorithms: A Multi-agent Feedback Control System

Tencent:

Link: https://meeting.tencent.com/dm/JfKIfn8oEUdC

ID:   804-513-122

Abstract：

Distributed algorithms have been playing an increasingly important role in many applications, such as machine learning, signal processing, and control. Significant research efforts have been devoted to developing and analyzing new algorithms for various applications. In this talk, we provide a fresh perspective to understand, analyze, and design distributed optimization algorithms through the lens of multi-rate feedback control. We show that a wide class of distributed algorithms, including popular decentralized/federated schemes, can be viewed as discretizing a certain continuous-time feedback control system,  possibly with multiple sampling rates. This key observation not only allows us to develop a generic framework to analyze the convergence of the entire algorithm class. More importantly, it also leads to an interesting way of designing new distributed algorithms.

Brief bio:

Xinwei Zhang is a Ph.D. in Electrical Engineering at the University of Minnesota. He received his Master's degree in  Electrical Engineering from the University of Minnesota and his Bachelor's degree in Automation from the University of Science and Technology of China. His research interest focuses on distributed optimization, non-convex optimization, federated learning, and differential privacy. His papers have been accepted by IEEE TSP, SIAM Journal on Optimization, ICML, ACC, and other transactions and conferences.