awjuliani/Meta-RL
TensorFlow implementation of Meta-RL A3C algorithm enabling reinforcement learning agents to adapt to new tasks without retraining.
Not currently ranked — collecting fresh signals.
star history
This repository provides a TensorFlow implementation of the Meta-Reinforcement Learning A3C algorithm based on the paper Learning to Reinforce Learn. It includes Jupyter notebooks demonstrating meta-learning across multiple task types: independent and dependent bandit tasks, restless bandits, contextual bandits using color-based reward signals, and a Rainbow Gridworld task where goal colors are randomized each episode requiring agents to learn on the fly.
Frequently asked
- What is awjuliani/Meta-RL?
- TensorFlow implementation of Meta-RL A3C algorithm enabling reinforcement learning agents to adapt to new tasks without retraining.
- Is Meta-RL open source?
- Yes — awjuliani/Meta-RL is open source, released under the MIT license.
- What language is Meta-RL written in?
- awjuliani/Meta-RL is primarily written in Jupyter Notebook.
- How popular is Meta-RL?
- awjuliani/Meta-RL has 407 stars on GitHub.
- Where can I find Meta-RL?
- awjuliani/Meta-RL is on GitHub at https://github.com/awjuliani/Meta-RL.