-
Machine Learning Research Lab, Volkswagen AG
- Munich, Germany
- https://icaruswizard.github.io/
- @xyzhang42
- in/xingyuan-zhang
- https://scholar.google.com/citations?user=BUqNLqcAAAAJ
Highlights
- Pro
Stars
[TMLR] Overcoming Knowledge Barriers: Online Imitation Learning from Visual Observation with Pretrained World Models
⚡⚡ Lightning Fast (~300TPS) Reinforcement Learning environment on latest Minecraft 🏝️
New repo collection for NVIDIA Cosmos: https://github.com/nvidia-cosmos
Official repository of our paper "Constrained Latent Action Policies for Model-Based Offline Reinforcement Learning" (NeurIPS 2024)
Octo is a transformer-based robot policy trained on a diverse mix of 800k robot trajectories.
Code for TRANSDREAMER: REINFORCEMENT LEARNING WITH TRANSFORMER WORLD MODELS
📦 The official Nextcloud installation method. Provides easy deployment and maintenance with most features included in this one Nextcloud instance.
[NeurIPS 2024] GenRL: Multimodal-foundation world models enable grounding language and video prompts into embodied domains, by turning them into sequences of latent world model states. Latent state…
Processed / Cleaned Data for Paper Copilot
Imitation learning benchmark focusing on complex locomotion tasks using MuJoCo.
SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning
A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.
This repository implements teleoperation of the Unitree humanoid robot using XR Devices.
Fast and flexible image augmentation library. Paper about the library: https://www.mdpi.com/2078-2489/11/2/125
MyoSuite is a collection of environments/tasks to be solved by musculoskeletal models simulated with the MuJoCo physics engine and wrapped in the OpenAI gym API.
CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks
ExORL: Exploratory Data for Offline Reinforcement Learning
Official implementation of AIME-v2 from "Overcoming Knowledge Barriers: Online Imitation Learning from Observation with Pretrained World Models"
A web-based collaborative LaTeX editor
🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning
(Crafter + NetHack) in JAX. ICML 2024 Spotlight.
A full-featured utility for managing Windows Subsystem for Linux (WSL)
PyTorch code and models for V-JEPA self-supervised learning from video.