Robust Model-Based Reinforcement Learning with an Adversarial Auxiliary Model

TL;DR

This paper introduces a robust model-based reinforcement learning approach using an adversarial auxiliary model to improve policy robustness in high-dimensional control tasks without requiring parametric simulators.

Contribution

It proposes a novel adversarial auxiliary transition model within a robust MDP framework, enhancing policy robustness without additional environment constraints.

Findings

Improved robustness of policies in distorted MDPs.

Effective integration of adversarial auxiliary model into RL algorithm.

Enhanced performance on high-dimensional MuJoCo tasks.

Abstract

Reinforcement learning has demonstrated impressive performance in various challenging problems such as robotics, board games, and classical arcade games. However, its real-world applications can be hindered by the absence of robustness and safety in the learned policies. More specifically, an RL agent that trains in a certain Markov decision process (MDP) often struggles to perform well in nearly identical MDPs. To address this issue, we employ the framework of Robust MDPs (RMDPs) in a model-based setting and introduce a novel learned transition model. Our method specifically incorporates an auxiliary pessimistic model, updated adversarially, to estimate the worst-case MDP within a Kullback-Leibler uncertainty set. In comparison to several existing works, our work does not impose any additional conditions on the training environment, such as the need for a parametric simulator. To test the effectiveness of the proposed pessimistic model in enhancing policy robustness, we integrate it into a practical RL algorithm, called Robust Model-Based Policy Optimization (RMBPO). Our experimental results indicate a notable improvement in policy robustness on high-dimensional MuJoCo control tasks, with the auxiliary model enhancing the performance of the learned policy in distorted MDPs. We further explore the learned deviation between the proposed auxiliary world model and the nominal model, to examine how pessimism is achieved. By learning a pessimistic world model and demonstrating its role in improving policy robustness, our research contributes towards making (model-based) RL more robust.