Tai Hoang

fluidgym. Plug-and-Play Benchmarking of Reinforcement Learning Algorithms for Large-Scale Flow Control

claw-code. An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

stable-worldmodel. A platform for reproducible world model research and evaluation

hydra_cluster_example. Example Project to get you started on Hydra and Cluster deployment

SONAR. Official code repository for "SONAR: Long-Range Graph Propagation Through Information Waves" accepted at NeurIPS 2025

mango. MaNGO - Adaptable Graph Network Simulators via Meta-Learning

reward_surface_alr. Reward Surface for visualizing RL rewards

SWAN. Official reference implementation of our paper "On Oversquashing in Graph Neural Networks Through The Lens of Dynamical Systems" accepted at AAAI 2025

felupe. Open finite element infrastructure for nonlinear computational mechanics.

newton. An open-source, GPU-accelerated physics simulation engine built upon NVIDIA Warp, specifically targeting roboticists and simulation researchers.

hydra_list_sweeper. Implementation of a list sweeper: instead of the cartesian product, sweep over the zipped list

s2gnn. Spatio-Spectral Graph Neural Networks (S²GNN)

linoss. Oscillatory State-Space Models

UCNS3D. Unstructured Compressible Navier Stokes 3D code (UCNS3D)

mujoco_warp. GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware.

traveling-waves-integrate. Repository to create traveling waves integrate special information through time

PDEControlGym. This gym provides implementations of various PDEs for easy testing and comparison of data-driven and classical PDE control algorithms.

controlgym. Controlgym: Large-Scale Control Environments for Benchmarking Reinforcement Learning Algorithms

XLB. XLB: Accelerated Lattice Boltzmann (XLB) for Physics-based ML

warp. A Python framework for GPU-accelerated simulation, robotics, and machine learning.

PhiFlow. A differentiable PDE solving framework for machine learning

rl. A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

ponita. Fast, Expressive SE(n) Equivariant Networks through Weight-Sharing in Position-Orientation Space.

movement-primitive-diffusion. Movement Primitive Diffusion (MPD) is a diffusion-based imitation learning method for high-quality robotic motion generation that focuses on gentle manipulation of deformable objects.

fancy_gym. Fancy Gym: Unifying interface for various RL benchmarks with support for Black Box approaches.

clusterduck. clusterduck is a hydra launcher plugin for running jobs in batches on a SLURM cluster. It is intended for small tasks on clusters where jobs have exclusive access to a node, such that submitting a single task to a node would be wasteful.

IsaacLab. Unified framework for robot learning built on NVIDIA Isaac Sim

parllel. A modular, flexible framework for developing performant algorithms in Reinforcement Learning.

MP_PyTorch. Movement Primitives in PyTorch

hyper-graph-nets. Implementation of different remote message passing strategies in graph neural networks for mesh-based physical simulation.

deep-variational-bayes-filters. Python

RoMBRL. Code for Bayes-Adaptive Deep Model-Based Policy Optimisation

GrMPN. Graph-based Motion Planning Networks

AVIAD_AVIJST. Towards Autoencoding Variational Inference for Aspect-based Opinion Summary

drone-hardware. Robust Quadrotor Frame for Machine Learning and Control Research

pysgmcmc. Bayesian Deep Learning with Stochastic Gradient MCMC Methods

piper. C++

PythonRobotics. Python sample codes and textbook for robotics algorithms.

PyAdvancedControl. Python codes for advanced control

gym-extensions. This repo is intended as an extension for OpenAI Gym for auxiliary tasks (multitask learning, transfer learning, inverse reinforcement learning, etc.)

gps. Guided Policy Search

turtlebot_mpepc. Reimplement MPEPC from MattDerry using MPEPC technique presented in Park's paper.

srl_global_planner. The SRL_GLOBAL_PLANNER package provides an implementation of the sampling based motion planners (RRT, RGG, RRT*, Theta*-RRT) as global planner plugin for Move-base, a ROS framework

gps. Guided Policy Search