Fault-Tolerant Federated Reinforcement Learning with Theoretical Guarantee
This addresses reliability and security issues in multi-agent decision-making systems, offering a foundational improvement for fault-tolerant FRL.
The paper tackles the lack of theoretical guarantees and fault tolerance in Federated Reinforcement Learning (FRL) by proposing a framework that ensures convergence and tolerates failures or attacks from less than half of the agents, with empirical validation on RL benchmarks.
The growing literature of Federated Learning (FL) has recently inspired Federated Reinforcement Learning (FRL) to encourage multiple agents to federatively build a better decision-making policy without sharing raw trajectories. Despite its promising applications, existing works on FRL fail to I) provide theoretical analysis on its convergence, and II) account for random system failures and adversarial attacks. Towards this end, we propose the first FRL framework the convergence of which is guaranteed and tolerant to less than half of the participating agents being random system failures or adversarial attackers. We prove that the sample efficiency of the proposed framework is guaranteed to improve with the number of agents and is able to account for such potential failures or attacks. All theoretical results are empirically verified on various RL benchmark tasks.