Stars
A framework for creating rich, 3D, Minecraft-like single and multi-agent environments for AI research. (Accepted at ICML 2025).
⚡⚡ Lightning Fast (~300TPS) Reinforcement Learning environment on latest Minecraft 🏝️
Re-implementation of pi0 vision-language-action (VLA) model from Physical Intelligence
FOCUS: An object-centric world implementation that allows for extracting per-object latent representations
[CoRL 2024] Open-TeleVision: Teleoperation with Immersive Active Visual Feedback
[ECCV2024] Video Foundation Models & Data for Multimodal Understanding
VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling
Implementation of Dreamer v3 in pytorch.
Playing Pokemon Red with Reinforcement Learning
Equivariant Steerable CNNs Library for Pytorch https://quva-lab.github.io/escnn/
PyTorch implementation of Soft Actor-Critic (SAC)
Code for "Temporal Difference Learning for Model Predictive Control"
A pattern-based approach for learning technical interview questions
Curiosity-driven Exploration by Self-supervised Prediction
Rainbow: Combining Improvements in Deep Reinforcement Learning
Implementation of random network distillation. paper link: https://arxiv.org/abs/1810.12894
robosuite: A Modular Simulation Framework and Benchmark for Robot Learning
A ROS/Python API for controlling and managing the Franka Emika Panda robot (real and simulated).
A large-scale benchmark and learning environment.
A platform for Reasoning systems (Reinforcement Learning, Contextual Bandits, etc.)
A toolkit for reproducible reinforcement learning research.
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
ChainerRL is a deep reinforcement learning library built on top of Chainer.