Certified Adversarial Robustness for Deep Reinforcement Learning

Abstract

Deep Neural Network-based systems are now the state-of-the-art in many robotics tasks, but their application in safety-critical domains remains dangerous without formal guarantees on network robustness. Small perturbations to sensor inputs (from noise or adversarial examples) are often enough to change network-based decisions, which was already shown to cause an autonomous vehicle to swerve into oncoming traffic. In light of these dangers, numerous algorithms have been developed as defensive mechanisms from these adversarial inputs, some of which provide formal robustness guarantees or certificates. This work leverages research on certified adversarial robustness to develop an online certified defense for deep reinforcement learning algorithms. The proposed defense computes guaranteed lower bounds on state-action values during execution to identify and choose the optimal action under a worst-case deviation in input space due to possible adversaries or noise. The approach is demonstrated on a Deep Q-Network policy and is shown to increase robustness to noise and adversaries in pedestrian collision avoidance scenarios and a classic control task.

arXiv commentPublished at Conference on Robot Learning (CoRL) 2019; (v2) contains minor updates to related works; (v3) acknowledged AWS

References in corpus (15)

• Practical Bayesian Optimization of Machine Learning Algorithms
• Ensemble Adversarial Training: Attacks and Defenses
• Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
• Certified Defenses against Adversarial Examples
• Robust Adversarial Reinforcement Learning
• Adversarial Risk and the Dangers of Evaluating Against Weak Attacks
• Efficient Neural Network Robustness Certification with General Activation Functions
• Evaluating Robustness of Neural Networks with Mixed Integer Programming
• Safe Model-based Reinforcement Learning with Stability Guarantees
• An approach to reachability analysis for feed-forward ReLU neural networks
• Uncertainty-Aware Reinforcement Learning for Collision Avoidance
• Efficient Formal Safety Analysis of Neural Networks
• Delving into adversarial attacks on deep policies
• Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential Prediction
• Online Robustness Training for Deep Reinforcement Learning

Cited by in corpus (4)

• Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations
• A Survey of Deep Reinforcement Learning in Recommender Systems: A Systematic Review and Future Directions
• Solving Dual Sourcing Problems with Supply Mode Dependent Failure Rates
• Certifiable Robustness to Adversarial State Uncertainty in Deep Reinforcement Learning