Behavior Regularized Offline Reinforcement Learning
This work addresses the challenge of applying RL in real-world scenarios with limited environment access, though it is incremental as it builds on existing methods to provide empirical insights.
The paper tackles the problem of offline reinforcement learning, where algorithms must learn from a fixed dataset without online interaction, by introducing the BRAC framework to evaluate methods on continuous control tasks. The result shows that many recent technical complexities are unnecessary for strong performance, with specific design choices identified as critical.
In reinforcement learning (RL) research, it is common to assume access to direct online interactions with the environment. However in many real-world applications, access to the environment is limited to a fixed offline dataset of logged experience. In such settings, standard RL algorithms have been shown to diverge or otherwise yield poor performance. Accordingly, recent work has suggested a number of remedies to these issues. In this work, we introduce a general framework, behavior regularized actor critic (BRAC), to empirically evaluate recently proposed methods as well as a number of simple baselines across a variety of offline continuous control tasks. Surprisingly, we find that many of the technical complexities introduced in recent methods are unnecessary to achieve strong performance. Additional ablations provide insights into which design choices matter most in the offline RL setting.