Contextual Policy Transfer in Reinforcement Learning Domains via Deep Mixtures-of-Experts
This addresses the challenge of transferring policies from simulation to real-world settings with limited data, though it appears incremental as it builds on existing policy reuse frameworks.
The paper tackles the problem of transferring knowledge from model-based learners to model-free learners in reinforcement learning by introducing a deep mixture-of-experts model that learns state-dependent beliefs over source task dynamics to match target dynamics. The method demonstrates effectiveness on OpenAI-Gym benchmarks, showing robustness to estimation errors and compatibility with standard policy reuse frameworks.
In reinforcement learning, agents that consider the context, or current state, when selecting source policies for transfer have been shown to outperform context-free approaches. However, none of the existing approaches transfer knowledge contextually from model-based learners to a model-free learner. This could be useful, for instance, when source policies are intentionally learned on diverse simulations with plentiful data but transferred to a real-world setting with limited data. In this paper, we assume knowledge of estimated source task dynamics and policies, and common sub-goals but different dynamics. We introduce a novel deep mixture-of-experts formulation for learning state-dependent beliefs over source task dynamics that match the target dynamics using state trajectories collected from the target task. The mixture model is easy to interpret, demonstrates robustness to estimation errors in dynamics, and is compatible with most learning algorithms. We then show how this model can be incorporated into standard policy reuse frameworks, and demonstrate its effectiveness on benchmarks from OpenAI-Gym.