Advancing Robustness in Deep Reinforcement Learning with an Ensemble Defense Approach
By: Adithya Mohan , Dominik Rößle , Daniel Cremers and more
Potential Business Impact:
Protects self-driving cars from tricky computer tricks.
Recent advancements in Deep Reinforcement Learning (DRL) have demonstrated its applicability across various domains, including robotics, healthcare, energy optimization, and autonomous driving. However, a critical question remains: How robust are DRL models when exposed to adversarial attacks? While existing defense mechanisms such as adversarial training and distillation enhance the resilience of DRL models, there remains a significant research gap regarding the integration of multiple defenses in autonomous driving scenarios specifically. This paper addresses this gap by proposing a novel ensemble-based defense architecture to mitigate adversarial attacks in autonomous driving. Our evaluation demonstrates that the proposed architecture significantly enhances the robustness of DRL models. Compared to the baseline under FGSM attacks, our ensemble method improves the mean reward from 5.87 to 18.38 (over 213% increase) and reduces the mean collision rate from 0.50 to 0.09 (an 82% decrease) in the highway scenario and merge scenario, outperforming all standalone defense strategies.
Similar Papers
Enhancing Security in Deep Reinforcement Learning: A Comprehensive Survey on Adversarial Attacks and Defenses
Cryptography and Security
Makes smart machines safer from tricky attacks.
Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks
Machine Learning (CS)
Tricks robots into making bad choices.
Robust Driving Control for Autonomous Vehicles: An Intelligent General-sum Constrained Adversarial Reinforcement Learning Approach
Machine Learning (CS)
Makes self-driving cars safer from tricky computer attacks.