I am a fourth year Ph.D. student at Tsinghua Shenzhen International Graduate School (SIGS) of Tsinghua Unviversity, advised by Assoc. Prof. Wenbo Ding. I initially joined Tsinghua-Berkey Shenzhen Institute (TBSI) as a master’s student and transitioned to the Ph.D. track after three years of research (2021-2024). Prior to that, I received the B.Eng degree in communication engineering from Chien-Shiung Wu College at Southeast University, Nanjing, China, in 2021. I was honored to complete my undergraduate thesis under the supervision of Prof. Shi Jin.

My research interests lie in reinforcement learning, multi-agent systems, and large models. Currently, I am working on mean field approximation methods grounded in information geometry, with a focus on scalable control strategies for large-scale swarm systems. My long-term goal is to develop algorithms capable of controlling swarms of arbitrary size with strong theoretical guarantees.

I am passionate about addressing real-world challenges through interdisciplinary research. Throughout my academic journey, I have been driven by a desire to advance the frontiers of intelligent systems. Beyond my primary research, I am increasingly interested in cross-disciplinary collaboration, particularly in tackling high-dimensional and complex systems.

Feel free to explore my work, and I welcome discussions and collaborations with researchers from all fields!

Contact

• Email: tanghz24@mails.tsinghua.edu.cn; thz21@mails.tsinghua.edu.cn;
• Github: Huaze Tang’s main page;
• Wechat: QR code

News

• [2025/06] Our paper Distributional Decision Transformer: Risk-Sensitive Offline RL via Quantile-Based Critics and Stochastic Return get accepted by IROS 2025! See you in Hangzhou!
• [2025/04] Our work in meituan has been awarded as 卓越实践奖 (Excellent Practice Award) by meituan! Thanks to my labmate Chao Wang and Zhenpeng Shi, my mentor and colleague in meituan and my supervisor.
• [2025/01] Our paper Residual Kernel Policy Network: Enhancing Stability and Robustness in RKHS-Based Reinforcement Learning gets accepted as poster in ICLR 2025! See you in Singapore!
• [2025/01] Our paper Mean-Field Aided QMIX: A Scalable and Flexible Q-Learning Approach For Large-Scale Agent Groups gets accepted in ICASSP 2025. See you in Hyderabad!
• [2024/08] Transitioned to the Ph.D. track.
• [2024/04] Present my work M^3ARL: Moment-Embedded Mean-Field Multi-Agent Reinforcement Learning for Continuous Action Space at ICASSP 2024.
• [2024/01] Start research internship in Meituan.