Online Robustness Training for Deep Reinforcement Learning
arXiv:1911.00887
Abstract
In deep reinforcement learning (RL), adversarial attacks can trick an agent into unwanted states and disrupt training. We propose a system called Robust Student-DQN (RS-DQN), which permits online robustness training alongside Q networks, while preserving competitive performance. We show that RS-DQN can be combined with (i) state-of-the-art adversarial training and (ii) provably robust training to obtain an agent that is resilient to strong attacks during training and evaluation.
References in corpus (15)
- Playing Atari with Deep Reinforcement Learning
- Dueling Network Architectures for Deep Reinforcement Learning
- Rainbow: Combining Improvements in Deep Reinforcement Learning
- Understanding Adversarial Training: Increasing Local Stability of Neural Nets through Robust Optimization
- Defensive Distillation is Not Robust to Adversarial Examples
- Tactics of Adversarial Attack on Deep Reinforcement Learning Agents
- Training verified learners with learned verifiers
- Policy Distillation
- Adversarial Attacks on Neural Network Policies
- Whatever Does Not Kill Deep Reinforcement Learning, Makes It Stronger
- Online Robust Policy Learning in the Presence of Unknown Adversaries
- Adversary A3C for Robust Reinforcement Learning
- Deep Robust Kalman Filter
- Collaborative Deep Reinforcement Learning
- A Provable Defense for Deep Residual Networks
Cited by in corpus (7)
- Robust Reinforcement Learning on State Observations with Learned Optimal Adversary
- Certified Adversarial Robustness for Deep Reinforcement Learning
- Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations
- Robust Deep Reinforcement Learning through Adversarial Loss
- Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RL
- CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothing
- Adversary Agnostic Robust Deep Reinforcement Learning