Robust Deep Reinforcement Learning Through Adversarial Attacks and Training : A Survey
Lucas Schott, Joséphine Delas, Hatem Hajri, Elies Gherbi, Reda Yaich, Nora Cuppens, Frédéric Cuppens, Sylvain Lamprier · arXiv (Cornell University) · 2024
Deep Reinforcement Learning (DRL) is a subfield of machine learning for training autonomous agents that take sequential actions across complex environments. Despite its significant performance in well-known environments, it remains susceptible to minor condition variations, raising concerns about its reliability in real-world applications. To improve usability, DRL must demonstrate trustworthiness and robustness. A way to improve the robustness of DRL to unknown changes in the environmental conditions and possible perturbations is through Adversarial Training, by training the agent against well-suited adversarial attacks on the observations and the dynamics of the environment. Addressing this critical issue, our work presents an in-depth analysis of contemporary adversarial attack and training methodologies, systematically categorizing them and comparing their objectives and operational mechanisms.