Robust Visual Domain Randomization for Reinforcement Learning
arXiv:1910.10537
Abstract
Producing agents that can generalize to a wide range of visually different environments is a significant challenge in reinforcement learning. One method for overcoming this issue is visual domain randomization, whereby at the start of each training episode some visual aspects of the environment are randomized so that the agent is exposed to many possible variations. However, domain randomization is highly inefficient and may lead to policies with high variance across domains. Instead, we propose a regularization method whereby the agent is only trained on one variation of the environment, and its learned state representations are regularized during training to be invariant across domains. We conduct experiments that demonstrate that our technique leads to more efficient and robust learning than standard domain randomization, while achieving equal generalization scores.
arXiv commentAccepted at the BeTR-RL Workshop at ICLR 2020
References in corpus (12)
- Unsupervised Data Augmentation for Consistency Training
- Learning Dexterous In-Hand Manipulation
- A Study on Overfitting in Deep Reinforcement Learning
- Spectrally-normalized margin bounds for neural networks
- Assessing Generalization in Deep Reinforcement Learning
- Learning Invariant Feature Spaces to Transfer Skills with Reinforcement Learning
- Generalization and Regularization in DQN
- A Dissection of Overfitting and Generalization in Continuous Reinforcement Learning
- Transfer Learning for Related Reinforcement Learning Tasks via Image-to-Image Translation
- Network Randomization: A Simple Technique for Generalization in Deep Reinforcement Learning
- Reinforcement Learning for Pivoting Task
- Mutual Alignment Transfer Learning