Multi-Source Transfer Learning for Deep Model-Based Reinforcement Learning
This work addresses the problem of sample inefficiency in reinforcement learning for researchers and practitioners, though it appears incremental as it builds on existing transfer learning concepts.
The paper tackled the challenge of reducing environment interactions in reinforcement learning by developing modular multi-source transfer learning techniques that automatically extract useful information from source tasks, regardless of differences in state-action space and reward function, and validated these techniques with extensive cross-domain experiments for visual control.
A crucial challenge in reinforcement learning is to reduce the number of interactions with the environment that an agent requires to master a given task. Transfer learning proposes to address this issue by re-using knowledge from previously learned tasks. However, determining which source task qualifies as the most appropriate for knowledge extraction, as well as the choice regarding which algorithm components to transfer, represent severe obstacles to its application in reinforcement learning. The goal of this paper is to address these issues with modular multi-source transfer learning techniques. The proposed techniques automatically learn how to extract useful information from source tasks, regardless of the difference in state-action space and reward function. We support our claims with extensive and challenging cross-domain experiments for visual control.