Online Robustness Training for Deep Reinforcement Learning

Autor:	Fischer, Marc, Mirman, Matthew, Stalder, Steven, Vechev, Martin
Jazyk:	angličtina
Rok vydání:	2019
Předmět:	FOS: Computer and information sciences Computer Science - Machine Learning Statistics - Machine Learning Machine Learning (stat.ML) Machine Learning (cs.LG)
Popis:	In deep reinforcement learning (RL), adversarial attacks can trick an agent into unwanted states and disrupt training. We propose a system called Robust Student-DQN (RS-DQN), which permits online robustness training alongside Q networks, while preserving competitive performance. We show that RS-DQN can be combined with (i) state-of-the-art adversarial training and (ii) provably robust training to obtain an agent that is resilient to strong attacks during training and evaluation.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bb97a25dee5563e5ad005272d10bf81b http://arxiv.org/abs/1911.00887 Zobrazit plný text záznamu