Offline Meta Learning of Exploration

Ron Dorfman, Aviv Tamar

Keywords: Meta-RL, Offline RL, Bayesian RL

Abstract: Consider the following problem: given the complete training histories of $N$ conventional RL agents, trained on $N$ different tasks, design a meta-agent that can quickly maximize reward in a new, unseen task from the same task distribution. In particular, while each conventional RL agent explored and exploited its own different task, the meta-agent must identify regularities in the data that lead to effective exploration/exploitation in the unseen task. This meta-learning problem is an instance of a setting we term Offline Meta Reinforcement Learning (OMRL). To solve our challenge, we take a Bayesian RL (BRL) view, and seek to learn a Bayes-optimal policy from the offline data. We extend the recently proposed VariBAD BRL algorithm to the off-policy setting, and demonstrate learning of approximately Bayes-optimal exploration strategies from offline data using deep neural networks. For the particular problem described above, our method learns effective exploration behavior that is qualitatively different from the exploration used by any RL agent in the data. Furthermore, we find that when applied to the online meta-RL setting (agent simultaneously collects data and improves its meta-RL policy), our method is significantly more sample efficient than the state-of-the-art VariBAD.

One-sentence Summary: Given complete training histories of RL agents trained on $N$ different tasks, we train a meta-RL agent that quickly solves a new task from the same task distribution.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=JFUyUa5iUi