Robust Deep Reinforcement Learning Through Adversarial Attacks and Training : A Survey

no code implementations • 1 Mar 2024 • Lucas Schott, Josephine Delas, Hatem Hajri, Elies Gherbi, Reda Yaich, Nora Boulahia-Cuppens, Frederic Cuppens, Sylvain Lamprier

Deep Reinforcement Learning (DRL) is an approach for training autonomous agents across various complex environments.

Adversarial Attack reinforcement-learning

Improving Robustness of Deep Reinforcement Learning Agents: Environment Attack based on the Critic Network

2 code implementations • 7 Apr 2021 • Lucas Schott, Hatem Hajri, Sylvain Lamprier

Existing approaches of the literature to generate meaningful disturbances of the environment are adversarial reinforcement learning methods.

Adversarial Attack reinforcement-learning +1

Stochastic sparse adversarial attacks

3 code implementations • 24 Nov 2020 • Manon Césaire, Lucas Schott, Hatem Hajri, Sylvain Lamprier, Patrick Gallinari

This paper introduces stochastic sparse adversarial attacks (SSAA), standing as simple, fast and purely noise-based targeted and untargeted attacks of neural network classifiers (NNC).