Robust Adversarial Reinforcement Learning via Bounded Rationality Curricula

TL;DR

This paper introduces Quantal Adversarial RL (QARL), a method that uses entropy regularization and bounded rationality to improve robustness and training efficiency in adversarial reinforcement learning tasks.

Contribution

It proposes a novel entropy-regularized approach that models agents with bounded rationality via Quantal Response Equilibrium, enabling curriculum learning of adversary rationality.

Findings

QARL outperforms RARL and baselines in MuJoCo tasks.

Modulating rationality via temperature improves training stability.

QARL enhances robustness against adversarial attacks.

Abstract

Robustness against adversarial attacks and distribution shifts is a long-standing goal of Reinforcement Learning (RL). To this end, Robust Adversarial Reinforcement Learning (RARL) trains a protagonist against destabilizing forces exercised by an adversary in a competitive zero-sum Markov game, whose optimal solution, i.e., rational strategy, corresponds to a Nash equilibrium. However, finding Nash equilibria requires facing complex saddle point optimization problems, which can be prohibitive to solve, especially for high-dimensional control. In this paper, we propose a novel approach for adversarial RL based on entropy regularization to ease the complexity of the saddle point optimization problem. We show that the solution of this entropy-regularized problem corresponds to a Quantal Response Equilibrium (QRE), a generalization of Nash equilibria that accounts for bounded rationality, i.e., agents sometimes play random actions instead of optimal ones. Crucially, the connection between the entropy-regularized objective and QRE enables free modulation of the rationality of the agents by simply tuning the temperature coefficient. We leverage this insight to propose our novel algorithm, Quantal Adversarial RL (QARL), which gradually increases the rationality of the adversary in a curriculum fashion until it is fully rational, easing the complexity of the optimization problem while retaining robustness. We provide extensive evidence of QARL outperforming RARL and recent baselines across several MuJoCo locomotion and navigation problems in overall performance and robustness.