schmidtdominik/Rainbow

Rainbow DQN implementation accompanying the paper "Fast and Data-Efficient Training of Rainbow" which reaches 205.7 median HNS after 10M frames. 🌈

34
/ 100
Emerging

This project helps machine learning researchers and reinforcement learning engineers train agents for game environments more efficiently. It takes raw game data from environments like Atari, Gym, or Procgen, processes it, and outputs a trained agent capable of playing these games. This is ideal for those focused on developing and evaluating advanced reinforcement learning algorithms with reduced computational demands.

No commits in the last 6 months.

Use this if you are a researcher or engineer looking to quickly train and benchmark Rainbow DQN agents on game environments with significantly less data and computational time.

Not ideal if you are a beginner looking for a simple, out-of-the-box solution for basic reinforcement learning tasks without prior experience in deep reinforcement learning.

reinforcement-learning deep-learning-research game-AI algorithm-benchmarking agent-training
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

44

Forks

4

Language

Python

License

MIT

Last pushed

Dec 11, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/schmidtdominik/Rainbow"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.