Welcome to AIR-DREAM (Decision-making Research for Empowered AI Methods) Lab Our research focus is to develop advanced learning-based data-driven decision-making theories and practical technologies that are robust, generalizable, and deployable to tackle real-world challenges. We work on fundamental learning algorithms, robust robotic control methods, optimization technologies for real-world AIoT systems, and data-driven decision-making tools & libraries.

AIR-DREAM Lab is a research group at Institute for AI Industry Research (AIR), Tsinghua University.

News

Dec. 2024: Our recent paper “Are Expressive Models Truly Necessary for Offline RL?” has been accepted in AAAI 2025!
Sep. 2024: Our two recent papers “Instruction-Guided Visual Masking” and “Diffusion-DICE: In-Sample Diffusion Guidance for Offline Reinforcement Learning” have been accepted in NeurIPS 2024!
Jul. 2024: Our two recent papers “DecisionNCE: Embodied Multimodal Representations via Implicit Preference Learning” and “Instruction-Guided Visual Masking” have won the Outstanding Paper Awards at ICML 2024 Workshop on Multi-modal Foundation Model meets Embodied AI (MFM-EAI).
May. 2024: Our four recent papers “DecisionNCE: Embodied Multimodal Representations via Implicit Preference Learning”, “OMPO: A Unified Framework for Reinforcement Learning under Policy and Dynamics Shifts”, “Offline-Boosted Actor-Critic: Adaptively Blending Optimal Historical Behaviors in Deep Off-Policy RL”, “Seizing Serendipity: Exploiting the Value of Past Success in Off-Policy Actor-Critic” have been accepted in ICML 2024!
Apr. 2024: Our recent survey paper “A Comprehensive Survey of Cross-Domain Policy Transfer for Embodied Agents” has been accepted in IJCAI 2024.
Jan. 2024: Our four recent papers “Revealing the Mystery of Distribution Correction Estimation via Orthogonal-gradient Update”, “Safe Offline Reinforcement Learning with Feasibility-Guided Diffusion Model”, “Query-Policy Misalignment in Preference-Based Reinforcement Learning”, and “OpenChat: Advancing Open-source Language Models with Mixed-Quality Data” have been accepted in ICLR 2024!
Oct. 2023: We have released “Data-Driven Control Library (D2C)”, which provides an easy-to-use and comprehensive library for real-world data-driven control & decision-making problems! Project page available at https://github.com/AIR-DI/D2C.
Sep. 2023: We have released “OpenChat: Advancing Open-source Language Models with Mixed-Quality Data”, which uses ideas from offline RL to fine-tune open-source large language models! Project page available at https://github.com/imoneoi/openchat.
Sep. 2023: Our two recent papers “Look Beneath the Surface: Exploiting Fundamental Symmetry for Sample-Efficient Offline RL” and “Offline Multi-Agent Reinforcement Learning with Implicit Global-to-Local Value Regularization” have been accepted in NeurIPS 2023!

🔥 We are hiring: we are looking for postdocs and student interns. If you are interested in the research directions of data-driven decision-making, please feel free to contact us!

Researches

Data-Driven Decision-Making Algorithms

Developing high-performance, robust, generalizable, and deployable data-driven decision-making algorithms for real-world problems.

Learning-based Methods for Robotics & Autonomous Driving

Learning robust and generalizable control / decision-making policies for robotics and autonomous driving using real-world data.

Data-Driven Methods for Sustainable Industrial and AIoT Systems

Using advanced data-driven AI methods to optimize real-world complex industrial and AIoT systems.

Tools & Libraries

Tools & software libraries for data-driven decision-making problems.

Recent Publications

Quickly discover relevant content by filtering publications.

Guan Wang, Haoyi Niu, Jianxiong Li, Li Jiang, Jianming HU, Xianyuan Zhan (2024). Are Expressive Models Truly Necessary for Offline RL?. The 39th Annual AAAI Conference on Artificial Intelligence (AAAI) (This paper is also accepted in NuerIPS 2024 Workshop on Open-World Agents (OWA)).

PDF Cite Code Project Project

Liyuan Mao, Haoran Xu, Weinan Zhang, Xianyuan Zhan, Amy Zhang (2024). Diffusion-DICE: In-Sample Diffusion Guidance for Offline Reinforcement Learning. In the Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS 2024).

PDF Cite Code Project Website

Jinliang Zheng, Jianxiong Li, Sijie Cheng, Yinan Zheng, Jiaming Li, Jihao Liu, Yu Liu, Jingjing Liu, Xianyuan Zhan (2024). Instruction-Guided Visual Masking. In the Thirty-Eighth Conference on Neural Information Processing Systems (NeurIPS 2024) (Outstanding paper award of ICML 2024 MFM-EAI Workshop).

PDF Cite Code Project Project Website

Hanfei Geng, Yi Sun, Yuanzhe Li, Jichao Leng, Xiangyu Zhu, Xianyuan Zhan, Yuanchun Li, Feng Zhao, Yunxin Liu (2024). TESLA: Thermally Safe, Load-Aware, and Energy-Efficient Cooling Control System for Data Centers. 53rd International Conference on Parallel Processing (ICPP 2024).

Cite Project

Yu Luo, Fuchun Sun, Tianying Ji, Xianyuan Zhan (2024). Bidirectional-Reachable Hierarchical Reinforcement Learning with Mutually Responsive Policies. 1st Reinforcement Learning Conference (RLC 2024).

PDF Cite Project

Yu Luo, Tianying Ji, Fuchun Sun, Jianwei Zhang, Huazhe Xu, Xianyuan Zhan (2024). OMPO: A Unified Framework for RL under Policy and Dynamics Shifts. Forty-first International Conference on Machine Learning (ICML 2024) (oral).

PDF Cite Project

See all publications

Meet the Team

Research team members

Xianyuan Zhan

Faculty Member

Xiangyu Zhu

Research Staff

Jianxiong Li

PhD Student

Jinliang Zheng

PhD Student

Yinan Zheng

PhD Student

Haoyi Niu

Research Intern

Peng Cheng

Research Intern

Tianying Ji

Research Intern

Wenjia Zhang

Research Intern

Xiao Hu

Research Intern

Yu Luo

Research Intern

Ziteng He

Research Intern

Alumni

Haoran Xu

PhD student at UT Austin, USA

Huiling Qin

Faculty at Beijing Normal University, Zhuhai

Li Jiang

PhD student at McGill University

Liyuan Mao

Undergraduate student at Shanghai Jiao Tong University

Weiye Xi

PhD student at Columbia Business School, USA

Xiangsen Wang

AI Algorithm engineer at Baidu

Yiwen Qiu

Master student at Carnegie Mellon University, USA

Yong Wang

PhD student at BIT (Beijing Institute of Technology)

Gallery

Contact

zhanxianyuan@mail.tsinghua.edu.cn
12 / F, block C, Qidi science and technology building, Tsinghua Science and Technology Park, Haidian District, Beijing, 100000
Find us on GitHub