Skip to yearly menu bar Skip to main content


Byzantine-Robust Online and Offline Distributed Reinforcement Learning

Yiding Chen · Xuezhou Zhang · Kaiqing Zhang · Mengdi Wang · Xiaojin Zhu

Auditorium 1 Foyer 50

Abstract: We consider a distributed reinforcement learning setting where multiple agents separately explore the environment and communicate their experiences through a central server. However, $\alpha$-fraction of agents are adversarial and can report arbitrary fake information. Critically, these adversarial agents can collude and their fake data can be of any sizes. We desire to robustly identify a near-optimal policy for the underlying Markov decision process in the presence of these adversarial agents. Our main technical contribution is \textsc{COW}, a novel algorithm for the \textit{robust mean estimation from batches} problem, that can handle arbitrary batch sizes. Building upon this new estimator, in the offline setting, we design a Byzantine-robust distributed pessimistic value iteration algorithm; in the online setting, we design a Byzantine-robust distributed optimistic value iteration algorithm. Both algorithms obtain near-optimal sample complexities and achieve superior robustness guarantee than prior works.

Live content is unavailable. Log in and register to view live content