Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks
This addresses the problem of deploying DRL in real-world robotics, though it is incremental as it builds on existing adversarial attack methods.
The paper tackles the vulnerability of deep reinforcement learning (DRL) in robotics to environmental perturbations by proposing the Adaptive Gradient-Masked Reinforcement (AGMR) Attack, which outperforms state-of-the-art methods in degrading agent performance and enhances robustness through adversarial defense.
Deep reinforcement learning (DRL) has emerged as a promising approach for robotic control, but its realworld deployment remains challenging due to its vulnerability to environmental perturbations. Existing white-box adversarial attack methods, adapted from supervised learning, fail to effectively target DRL agents as they overlook temporal dynamics and indiscriminately perturb all state dimensions, limiting their impact on long-term rewards. To address these challenges, we propose the Adaptive Gradient-Masked Reinforcement (AGMR) Attack, a white-box attack method that combines DRL with a gradient-based soft masking mechanism to dynamically identify critical state dimensions and optimize adversarial policies. AGMR selectively allocates perturbations to the most impactful state features and incorporates a dynamic adjustment mechanism to balance exploration and exploitation during training. Extensive experiments demonstrate that AGMR outperforms state-of-the-art adversarial attack methods in degrading the performance of the victim agent and enhances the victim agent's robustness through adversarial defense mechanisms.