Toward Evaluating Robustness Of Reinforcement Learning With Adversarial Policy

Abstract

Reinforcement learning agents are susceptible to evasion attacks during deployment. In single-agent environments, these attacks can occur through imperceptible perturbations injected into the inputs of the victim policy network. In multi-agent environments, an attacker can manipulate an adversarial opponent to influence the victim policy's observations indirectly. While adversarial policies offer a promising technique to craft such attacks, current methods are either sample-inefficient due to poor exploration strategies or require extra surrogate model training under the black-box assumption. To address these challenges, in this paper, we propose Intrinsically Motivated Adversarial Policy (IMAP) for efficient black-box adversarial policy learning in both single- and multi-agent environments. We formulate four types of adversarial intrinsic regularizers -- maximizing the adversarial state coverage, policy coverage, risk, or divergence -- to discover potential vulnerabilities of the vict

Stats

• citations3
• S2 citations—
• github stars0
• HF likes0
• heat score4.52
• arxiv keyzheng2023toward

Related papers

• Adversarial Policies: Attacking Deep Reinforcement Learning (2019)0.00
• Robust Deep Reinforcement Learning Against Adversarial Behavior Manipulation (2024)0.00
• Constrained Black-box Attacks Against Cooperative Multi-agent Reinforcement Learning (2025)0.00
• Online Robust Policy Learning In The Presence Of Unknown Adversaries (2018)0.00
• Imitating Opponent To Win: Adversarial Policy Imitation Learning In Two-player Competitive Games (2022)0.00
• SUB-PLAY: Adversarial Policies Against Partially Observed Multi-agent Reinforcement Learning Systems (2024)0.00
• Attacking And Defending Deep Reinforcement Learning Policies (2022)0.00
• Towards Robust Policy: Enhancing Offline Reinforcement Learning With Adversarial Attacks And Defenses (2024)3.58