Federated Reinforcement Learning with Environment Heterogeneity
This addresses the challenge of collaborative learning in distributed systems with varying environments, which is incremental as it builds on existing federated learning and RL methods.
The paper tackles the problem of federated reinforcement learning with environment heterogeneity, where agents learn a single policy without sharing trajectories, and proposes two algorithms (QAvg and PAvg) that converge to suboptimal solutions with performance dependent on heterogeneity, along with a personalization heuristic that improves training and generalization.
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.