Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks
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.
Code (0)
등록된 구현이 없습니다.
Tasks
Adversarial AttackAdversarial DefenseDeep Reinforcement LearningSimilar Papers 제목 키워드 기반
Robust Reinforcement Learning as a Stackelberg Game via Adaptively-Regularized Adversarial Training
Robust Reinforcement Learning (RL) focuses on improving performances under model errors or adversarial attacks, which facilitates the real-life deployment of RL agents. Robust Adversarial Reinforcement Learning (RARL) is…
reinforcement-learningReinforcement LearningReinforcement Learning (RL)Gaussian Splatting Under Attack: Investigating Adversarial Noise in 3D Objects
3D Gaussian Splatting has advanced radiance field reconstruction, enabling high-quality view synthesis and fast rendering in 3D modeling. While adversarial attacks on object detection models are well-studied for 2D image…
Autonomous Drivingobject-detectionObject DetectionZero-Shot Object DetectionAdaptive Masked Proxies for Few-Shot Segmentation
Deep learning has thrived by training on large-scale datasets. However, in robotics applications sample efficiency is critical. We propose a novel adaptive masked proxies method that constructs the final segmentation lay…
Continual LearningFew-Shot Semantic SegmentationSegmentationSemantic Segmentation+2AMP: Adaptive Masked Proxies for Few-Shot Segmentation
Deep learning has thrived by training on large-scale datasets. However, in robotics applications sample efficiency is critical. We propose a novel adaptive masked proxies method that constructs the final segmentation lay…
Continual LearningFew-Shot Semantic SegmentationSegmentationSemantic SegmentationAdaMAE: Adaptive Masking for Efficient Spatiotemporal Learning with Masked Autoencoders
Masked Autoencoders (MAEs) learn generalizable representations for image, text, audio, video, etc., by reconstructing masked input data from tokens of the visible data. Current MAE approaches for videos rely on random pa…
Action ClassificationRepresentation Learning