denisyarats/drq
PyTorch implementation of DrQ, a data-regularized deep reinforcement learning algorithm for training control policies from pixel observations.
Not currently ranked — collecting fresh signals.
star history
This repository implements the DrQ algorithm from the paper Image Augmentation Is All You Need. It provides a model-free reinforcement learning agent using actor-critic methods with data augmentation for regularization. The implementation trains on pixel-based inputs using the DM Control and Gym environments with MuJoCo physics simulation, achieving state-of-the-art performance on continuous control tasks.
Frequently asked
- What is denisyarats/drq?
- PyTorch implementation of DrQ, a data-regularized deep reinforcement learning algorithm for training control policies from pixel observations.
- Is drq open source?
- Yes — denisyarats/drq is open source, released under the MIT license.
- What language is drq written in?
- denisyarats/drq is primarily written in Jupyter Notebook.
- How popular is drq?
- denisyarats/drq has 422 stars on GitHub.
- Where can I find drq?
- denisyarats/drq is on GitHub at https://github.com/denisyarats/drq.