About me

I am a Ph.D. candidate in Computer Science at University of Southern California. I am co-advised by Professor Bistra Dilkina and Professor Sven Koenig.

In 2019, I received my B.E. in computer science from Yao Class at Tsinghua University.

From May to December in 2022, I was a research scientist intern at Meta AI, FAIR, where I worked closely with Benoit Steiner and Yuandong Tian. In Summer 2021, I was a research scientist intern at Amazon Robotics where I worked with Vikas Shivashankar. In Spring 2018, I was a research intern at Carneige Mellon University advised by Professor Fei Fang. In Summer 2018, I was a research intern at The University of Hong Kong advised by Professor Zhiyi Huang.

My research interests lie in artificial intelligence, with a focus on multi-agent systems, combinatorial optimization, machine learning, computational game theory and computational sustainability.

CV

News

[2024.06] We released Distributional MIPLIB, the first multi-domain library of problem distributions for advancing machine learning-guided mixed integer linear programming solving.

• [2024.08] I defended my thesis on Improving Decision-Making in Search Algorithms with Machine Learning for Combinatorial Optimizations.
• [2024.07] Our paper on using contrastive learning for backdoor predictions for MILPs has been accepted to ECAI 2024.
• [2024.05] Our papers ConPaS and GenCo have been accepted to ICML 2024.
• [2023.12] Our work BALANCE, a bandit-based Large Neighborhood Search for MAPF, is accepted to AAAI 2024.
• [2023.10] I gave a research talk on neural combinatorial optimization at The D.E. Shaw Group.
• [2023.09] I passed my PhD thesis proposal!
• [2023.09] Our work with FAIR, LANCER: Learning Decision Losses for Mathematical Optimization Under Partial Information, is accepted to NeurIPS 2023!
• [2023.07] Our SurCo paper won the Outstanding Paper Award at the ICML 2023 SODS Workshop.
• [2023.07] Our paper on solving large scale planning problems in automated warehouses is named the Best Student Paper Runner-Up at ICAPS-2023.
• [2023.04] Our papers on ML-guided optimization for mixed integer programs and non-linear combinatorial optimization with Facebook AI Research are accepted to ICML-2023.
• [2023.03] Our works on using ML for combinatorial optimizations are presented at the IPAM Artificial Intelligence and Discrete Optimization Workshop. The recordings can be found on YouTube: Bistra’s presentation, Yuandong’s presentation. I also presented a poster which is a summary of my work on ML-guided multi-agent path finding.
• [2023.02] Our paper with Amazon Robotics on solving large-scale planning problems in automated warehouses is accepted to ICAPS-2023. I also presented this work at AAAI-23 Workshop on Multi-Agent Path Finding.
• [2023.01] Our paper with Facebook AI Research on applying large neighborhood search to solve integer linear programs is accepted to CPAIOR-2023. We discovered a simple yet effective destroy heuristic based on the LP relaxation of Local Branching.
• [2022.10] I presented my work on ML-guided multi-agent path finding at INFORMS Annual Meeting 2022.
• [2022.08] I started a new position as a student researcher at Meta AI, FAIR.