Model Imitation for Model-Based Reinforcement Learning
This addresses a key bottleneck in MBRL for improving sample efficiency in real-world applications, though it is an incremental improvement over existing methods.
The paper tackles the problem of model-based reinforcement learning (MBRL) suffering from distribution mismatches in long-horizon rollouts due to one-step transition model errors, which increases sample complexity. It proposes learning the transition model by matching multi-step rollout distributions using WGAN, theoretically minimizing cumulative reward differences and experimentally achieving competitive or superior sample complexity and average return compared to state-of-the-art methods.
Model-based reinforcement learning (MBRL) aims to learn a dynamic model to reduce the number of interactions with real-world environments. However, due to estimation error, rollouts in the learned model, especially those of long horizons, fail to match the ones in real-world environments. This mismatching has seriously impacted the sample complexity of MBRL. The phenomenon can be attributed to the fact that previous works employ supervised learning to learn the one-step transition models, which has inherent difficulty ensuring the matching of distributions from multi-step rollouts. Based on the claim, we propose to learn the transition model by matching the distributions of multi-step rollouts sampled from the transition model and the real ones via WGAN. We theoretically show that matching the two can minimize the difference of cumulative rewards between the real transition and the learned one. Our experiments also show that the proposed Model Imitation method can compete or outperform the state-of-the-art in terms of sample complexity and average return.