Han Zheng
I'm a Ph.D. student at MIT, affiliated with the Laboratory for Information and Decision Systems (LIDS), advised by Prof. Cathy Wu. Previously I was at the University of Michigan working with the late Prof. Huei Peng, and I hold dual B.S./B.E. degrees from UIUC and Zhejiang University.
Research
I'm interested in scalable planning and decision-making for multi-agent systems, combining reinforcement learning with classical search and combinatorial optimization.
arXiv, 2026
First framework for autonomous multi-agent evolution on open-ended problems. CORAL uses long-running LLM agents that explore, reflect, and collaborate through shared persistent memory and asynchronous execution, achieving 3–10× higher improvement rates over fixed evolutionary baselines across mathematical, algorithmic, and systems optimization tasks.
JAIR, 2026
Developed RL-RH-PP, combining deep reinforcement learning with search-based prioritized planning for
lifelong MAPF. Demonstrated superior throughput and strong generalization across agent densities and planning horizons.
INFORMS Data Mining, 2025
★ Best Paper Finalist
Self-improving LLM framework that builds an experience library for formulating optimization programs across diverse problem domains.
IROS, 2024
Pioneered BK-PBS, integrating offline-trained behavior prediction with multi-agent pathfinding to coordinate
connected and human-driven vehicles under realistic kinematic constraints on highways.
ICRA, 2024
Proposed OBS-KATS for signal-free intersection coordination. Proved soundness, completeness, and optimality
of the crossing order search with significant throughput improvements.
MECC, 2023
Hierarchical eco-driving algorithms for electric connected vehicles, reducing energy consumption near adaptive traffic signals in urban areas.
Industry
Research Intern — Amazon AGI
Incoming, 2026
Bellevue, WA
Reinforcement fine-tuning for Nova models.
Research Intern — Symbotic
May – Sep 2025
Wilmington, MA
Research on learning-based path planning and search algorithms for large-scale warehouse robot fleet coordination.
Integrated reinforcement learning and combinatorial optimization for routing efficiency and congestion management.
Education
2023 –
Massachusetts Institute of Technology
Ph.D. in Mechanical Engineering & Computational Science and Engineering
Ph.D. in Mechanical Engineering & Computational Science and Engineering
2021 – 2023
University of Michigan – Ann Arbor
Ph.D. in Mechanical Engineering (transferred to MIT)
Ph.D. in Mechanical Engineering (transferred to MIT)
2017 – 2021
University of Illinois at Urbana-Champaign
B.S. Mechanical Engineering with Highest Honors
B.S. Mechanical Engineering with Highest Honors
2017 – 2021
Zhejiang University
B.E. in Mechanical Engineering
B.E. in Mechanical Engineering