eloialonso/diamond

DIAMOND (DIffusion As a Model Of eNvironment Dreams) is a reinforcement learning agent trained in a diffusion world model. NeurIPS 2024 Spotlight.

45
/ 100
Emerging

This project offers a way to train and experiment with reinforcement learning agents within highly realistic simulated environments. You input an agent's policy and a pre-trained 'world model' (like for Atari games or CS:GO), and it outputs how the agent behaves and learns within that simulated world. Game AI researchers and developers can use this to develop and test AI behaviors more efficiently.

1,988 stars. No commits in the last 6 months.

Use this if you are a game AI researcher or developer looking to train and evaluate reinforcement learning agents in visually rich, simulated game environments.

Not ideal if you are not working with AI agents in game environments or are looking for a simple, off-the-shelf game AI solution without needing to train custom models.

game-AI reinforcement-learning AI-simulation world-modeling game-development
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

1,988

Forks

145

Language

Python

License

MIT

Last pushed

Dec 06, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/eloialonso/diamond"

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