A study of natural robustness of deep reinforcement learning algorithms towards adversarial perturbations
Date
2024-09-09
Authors
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Elsevier B.V. on behalf of KeAi Communications Co. Ltd
Abstract
Deep reinforcement learning (DRL) has been shown to have numerous potential applications in the real world. However, DRL algorithms are still extremely sensitive to noise and adversarial perturbations, hence inhibiting the deployment of RL in many real-life applications. Analyzing the robustness of DRL algorithms to adversarial attacks is an important prerequisite to enabling the widespread adoption of DRL algorithms. Common perturbations on DRL frameworks during test time include perturbations to the observation and the action channel. Compared with observation channel attacks, action channel attacks are less studied; hence, few comparisons exist that compare the effectiveness of these attacks in DRL literature. In this work, we examined the effectiveness of these two paradigms of attacks on common DRL algorithms and studied the natural robustness of DRL algorithms towards various adversarial attacks in hopes of gaining insights into the individual response of each type of algorithm under different attack conditions.
Series Number
Journal Issue
Is Version Of
Versions
Series
Academic or Administrative Unit
Type
article
Comments
This article is published as Liu, Qisai, Xian Yeow Lee, and Soumik Sarkar. "A study of natural robustness of deep reinforcement learning algorithms towards adversarial perturbations." AI Open 5 (2024): 126-141. doi: https://doi.org/10.1016/j.aiopen.2024.08.005.
Rights Statement
© 2024 The Authors. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).