Optimizing Dynamic Programming-Based Algorithms

Fast and simple implementation of RL algorithms, designed to run fully on GPU.

Imitation learning benchmark focusing on complex locomotion tasks using MuJoCo.

A library for differentiable nonlinear optimization

A library for differentiable robotics.

The official Meta Llama 3 GitHub site

Differentiable Rasteriser implemented in JAX. Reference: https://github.com/erwincoumans/tinyrenderer, https://github.com/ssloy/tinyrenderer/wiki; PR: https://github.com/google/brax/pull/367

Hamilton-Jacobi reachability analysis in JAX.

🚀 A fast safe reinforcement learning library in PyTorch

NeurIPS 2023: Safety-Gymnasium: A Unified Safe Reinforcement Learning Benchmark

[JMLR] OmniSafe is an infrastructural framework for accelerating SafeRL research.