Federated Reinforcement Learning with Environment Heterogeneity
Jin, Hao, Peng, Yang, Yang, Wenhao, Wang, Shusen, Zhang, Zhihua
We study a Federated Reinforcement Learning (FedRL) problem in which $n$ agents collaboratively learn a single policy without sharing the trajectories they collected during agent-environment interaction. We stress the constraint of environment heterogeneity, which means $n$ environments corresponding to these $n$ agents have different state transitions. To obtain a value function or a policy function which optimizes the overall performance in all environments, we propose two federated RL algorithms, \texttt{QAvg} and \texttt{PAvg}. We theoretically prove that these algorithms converge to suboptimal solutions, while such suboptimality depends on how heterogeneous these $n$ environments are. Moreover, we propose a heuristic that achieves personalization by embedding the $n$ environments into $n$ vectors. The personalization heuristic not only improves the training but also allows for better generalization to new environments.
Apr-6-2022
- Country:
- North America > United States
- Virginia (0.04)
- New Jersey > Mercer County
- Princeton (0.04)
- Massachusetts > Middlesex County
- Belmont (0.04)
- Europe > Spain
- Valencian Community > Valencia Province > Valencia (0.04)
- Asia > China
- North America > United States
- Genre:
- Research Report (0.64)
- Industry:
- Information Technology > Security & Privacy (0.46)
- Technology: