Simple single-file baselines for Q-Learning in pure-GPU setting

Simplifying reinforcement learning for complex game environments

The NetHack Learning Environment

RLeXplore provides stable baselines of exploration methods in reinforcement learning, such as intrinsic curiosity module (ICM), random network distillation (RND) and rewarding impact-driven explora…

A fast and simple perlin noise generator using numpy

Simple, open source, 2D impulse based physics engine for educational use.

Box2D is a 2D physics engine for games

HTML5 Port of Incredibots 2 CE

Official implementation of the RLC 2024 paper "Policy-Guided Diffusion"

A toolkit for developing and comparing reinforcement learning algorithms.

♟️ Vectorized RL game environments in JAX

Massively parallel rigidbody physics simulation on accelerator hardware.

Accelerated Quality-Diversity

🕹️ A diverse suite of scalable reinforcement learning environments in JAX

⚡ Flashbax: Accelerated Replay Buffers in JAX

Code for the paper Refining Minimax Regret for Unsupervised Environment Design (ICML 2024) https://arxiv.org/abs/2402.12284

MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Research

🏛️A research-friendly codebase for fast experimentation of single-agent reinforcement learning in JAX • End-to-End JAX RL

Efficient baselines for autocurricula in JAX.

Benchmarking the Spectrum of Agent Capabilities

Multi-Agent Reinforcement Learning with JAX

Animation engine for explanatory math videos

Long-Term Evolution Project of Reinforcement Learning

SkillHack: A Benchmark for Skill Transfer in Open-Ended Reinforcement Learning

(Crafter + NetHack) in JAX. ICML 2024 Spotlight.