Reinforcement Learning Experience Reuse with Policy Residual Representation
This addresses sample efficiency in reinforcement learning for domains like navigation and gaming, but it is incremental as it builds on existing experience reuse methods.
The paper tackles the problem of representing and reusing experience in reinforcement learning by proposing a policy residual representation (PRR) network that extracts multiple granularities of experience, and it shows that PRR outperforms state-of-the-art approaches in grid world, locomotion, and video game tasks.
Experience reuse is key to sample-efficient reinforcement learning. One of the critical issues is how the experience is represented and stored. Previously, the experience can be stored in the forms of features, individual models, and the average model, each lying at a different granularity. However, new tasks may require experience across multiple granularities. In this paper, we propose the policy residual representation (PRR) network, which can extract and store multiple levels of experience. PRR network is trained on a set of tasks with a multi-level architecture, where a module in each level corresponds to a subset of the tasks. Therefore, the PRR network represents the experience in a spectrum-like way. When training on a new task, PRR can provide different levels of experience for accelerating the learning. We experiment with the PRR network on a set of grid world navigation tasks, locomotion tasks, and fighting tasks in a video game. The results show that the PRR network leads to better reuse of experience and thus outperforms some state-of-the-art approaches.