Byzantine-Robust Gossip: Insights from a Dual Approach

arXiv:2405.03449v1 Announce Type: new
Abstract: Distributed approaches have many computational benefits, but they are vulnerable to attacks from a subset of devices transmitting incorrect information. This paper investigates Byzantine-resilient algorithms in a decentralized setting, where devices communicate directly with one another. We leverage the so-called dual approach to design a general robust decentralized optimization method. We provide both global and local clipping rules in the special case of average consensus, with tight convergence guarantees. These clipping rules are practical, and …

abstract algorithms arxiv attacks benefits computational cs.lg decentralized design devices distributed general information insights paper resilient robust stat.ml type vulnerable