arXiv 2026
HARBOR: A Harness Framework for Agentic Robot Reinforcement Learning
Automates the end-to-end robot RL workflow, from package installation to policy tuning.
Publications
15* equal contribution
2026
IROS 2026
Scalable Multi-Task Data Generation via Reinforcement Learning for Language-Conditioned Bimanual Dexterous Manipulation
Generates multi-task demonstrations for language-conditioned bimanual manipulation via RL.
2 more from 2026Show fewer
arXiv 2026
Nautilus: From One Prompt to Plug-and-Play Robot Learning
Turns a single prompt into reproduction, evaluation, fine-tuning, and deployment workflows.
ICRA 2026
SE(3)-PoseFlow: Estimating 6D Pose Distributions for Uncertainty-Aware Robotic Manipulation
Estimates full 6D pose distributions via flow matching on SE(3) for uncertainty-aware manipulation.
2025
CoRL 2025
Morphologically Symmetric Reinforcement Learning for Ambidextrous Bimanual Manipulation
Exploits morphological symmetry to learn ambidextrous bimanual manipulation.
ICML 2025
DIME: Diffusion-Based Maximum Entropy Reinforcement Learning
Trains diffusion policies with maximum-entropy RL, reaching state-of-the-art returns.
2024
NeurIPS 2024
Learning Multimodal Behaviors from Scratch with Diffusion Policy Gradient
Learns multimodal diffusion policies from scratch, without demonstrations.
RSS 2024
Reconciling Reality through Simulation: A Real-to-Sim-to-Real Approach for Robust Manipulation
Robustifies real-world policies by fine-tuning them in digital twins built on the fly.
2023
ICML 2023
Parallel Q-Learning: Scaling Off-policy Reinforcement Learning under Massively Parallel Simulation
Scales off-policy RL to 10,000+ parallel environments.
2 more from 2023Show fewer
Games 2023
Social Learning for Sequential Driving Dilemmas
Identifies when social dilemmas arise in autonomous vehicles’ sequential decisions.
ICLR Workshop 2023
Stationary Deep Reinforcement Learning with Quantum K-spin Hamiltonian Equation
Adds a quantum K-spin Hamiltonian regularizer that steers policies toward better optima.
2022
NeurIPS 2022
Homomorphic Matrix Completion
Achieves exact matrix completion under differential privacy, at the cost of more samples.
1 more from 2022Show fewer
IEEE IV 2022
Social Learning In Markov Games: Empowering Autonomous Driving
Studies how autonomous vehicles form social norms through RL in Markov games.
2021
Show 2 papersShow fewer
ICAIF 2021
FinRL-Podracer: High Performance and Scalable Deep Reinforcement Learning for Quantitative Finance
Trains a NASDAQ-100 trading agent in 10 minutes on 80 A100 GPUs.
NeurIPS Workshop 2021
ElegantRL-Podracer: Scalable and Elastic Library for Cloud-native Deep Reinforcement Learning
Scales cloud-native DRL training across millions of GPU cores.
Open Source Projects
2
GitHub16.2k
FinRL: Financial Reinforcement Learning
Brings deep reinforcement learning to quantitative finance and automated trading.
GitHub4.4k
ElegantRL “小雅”: Massively Parallel Library for Cloud-native Deep Reinforcement Learning
Runs massively parallel DRL training on cloud-native infrastructure.
As a leader of this project, I have been contributing to
- develop a series of large-scale training frameworks,
- implemente SOTA algorithms and techniques,
- build the documentation website.
Starting from Mar. 2021, I started to write tutorial blogs for the community,
- ElegantRL: Much More Stable Deep Reinforcement Learning Algorithms than Stable-Baseline3MLearning.ai, Mar. 3, 2022
- ElegantRL-Podracer: A Scalable and Elastic Library for Cloud-Native Deep Reinforcement LearningTowards data science, Dec. 11, 2021
- ElegantRL: Mastering PPO AlgorithmsTowards data science, May. 3, 2021
- ElegantRL Demo: Stock Trading Using DDPG (Part II)MLearning.ai, Apr. 19, 2021
- ElegantRL Demo: Stock Trading Using DDPG (Part I)MLearning.ai, Mar. 28, 2021
- ElegantRL-Helloworld: A Lightweight and Stable Deep Reinforcement Learning LibraryTowards data science, Mar. 4, 2021
Book Chapter
1
Elsevier 2021
High-performance Tensor Decompositions for Compressing and Accelerating Deep Neural Networks
Compresses and accelerates neural networks with high-performance tensor decompositions.