About Me


I am an assistant professor in Computer Science and Data Science at UChicago, the director of the SIGMA Lab (Strategic IntelliGence for Machine Agents), an AI2050 early career fellow, and a (part-time) staff scientist at Google Research.



At UChicago, I enjoy working with -- and drawing inspiration from -- a group of incredible students. Over the years, I was fortunate to have also worked closely with many exceptional (by-then) PhD students outside UChi, such as Tao Lin, Yang Yu, Junjie Chen, Thomas K. Buening, Konstantin Zabarnyi, Wei Tang, Huazheng Wang, and Chuanhao Li.


Interested in joining our research lab? Please read this.


Contact


Office: Room 312, 5460 S. University Ave

E-mail: haifengxu AT uchicago DOT edu


About My Research


I work on AI agents and data-driven decision making, particularly in multi-agent environments. The overarching goal is to go beyond recognition-level capabilities of AI/ML, and develop "agency-level" AI capabilities, such as intelligent communications, reasoning, decisions, interactions, etc. Large data-driven learning paradigms are foundational for these, but seem not sufficient according to evidences thus far. The integration of data-driven methods with more structured methodologies (such as strategic reasoning, optimization, behavioral modeling and knowledge graph) seems essential for next breakthroughs. To push this frontier, our lab has been actively working on the following research directions. Please see our lab website for more details.

  1. Rethinking online contents and GenAI through the lens of computational economics

    2. Representative Recent Papers (see Schmidt Sciences' interview and my talk at Sigecom Winter25 meeting):

    1. Mechanism Design for Large Language Models

      (α-β) Paul Duetting, Vahab Mirrokni, Renato Paes Leme, Haifeng Xu, and Song Zuo

      WWW 2024: Proc. ACM Web Conference, 2024 (Best Paper Award)

      Invited for Highlights Beyond EC at EC 2024 and IJCAI'25 Sister Conferences' Best Paper Track


    2. Human vs. Generative AI in Content Creation Competition: Symbiosis or Conflict?

      Fan Yao, Chuanhao Li, Denis Nekipelov, Hongning Wang and Haifeng Xu

      ICML 2024: Proc. 41th International Conference on Machine Learning, 2024


    3. Rethinking Incentives in Recommender Systems: Are Monotone Rewards Always Beneficial?

      Fan Yao, Chuanhao Li, Karthik Abinav Sankararaman, Yiming Liao, Yan Zhu, Qifan Wang, Hongning Wang and Haifeng Xu

      NeurIPS 2023: Proc. 37th Conference on Neural Information Processing Systems, 2023.

      Algorithms based on this work are live tested on Instagram (see this KDD'24 paper for reported results)


    4. How Bad is Top-K Recommendation under Competing Content Creators?

      Fan Yao, Chuanhao Li, Denis Nekipelov, Hongning Wang and Haifeng Xu

      ICML 2023: Proc. 40th International Conference on Machine Learning, 2023 (oral presentation, 2.3%)


  2. Information design, principal-agency and foundations of intelligent agent communications

    3. Representative Recent Papers (see a tutorial organized by my student Jibang at WINE24):

    1. Algorithmic Cheap Talk

      (α-β) Yakov Babichenko, Inbal Talgam-Cohen, Haifeng Xu and Konstantin Zabarnyi

      EC 2024: Proc. 25th ACM Conference on Economics and Computation, 2024.


    2. Learning to Persuade on the Fly: Robustness Against Ignorance

      You Zu, Krishnamurthy Iyer, Haifeng Xu

      Operations Research, 2024 (also appears at EC 2021).


    3. Robust Stackelberg Equilibria

      (α-β) Jiarui Gan, Minbiao Han, Jibang Wu, and Haifeng Xu

      EC 2023: Proc. 24th ACM Conference on Economics and Computation, 2023.

      R&R under Mathematical Programming


    4. Sequential Information Design: Markov Persuasion Process and Its Efficient Reinforcement Learning

      Jibang Wu, Zixuan Zhang, Zhe Feng, Zhaoran Wang, Zhuoran Yang, Michael I. Jordan and Haifeng Xu

      EC 2022: Proc. 23th ACM Conference on Economics and Computation, 2022.


    5. On the Tractability of Public Persuasion with No Externalities

      Haifeng Xu

      SODA 2020: Proc. 31st ACM-SIAM Symposium on Discrete Algorithms, 2020.



  3. Valuation and pricing of information/data

    4. Representative Recent Papers (see also my tutorial at AAAI23):

    1. Selling Data To a Machine Learner: Pricing via Costly Signaling

      (α-β) Junjie Chen, Minming Li and Haifeng Xu

      ICML 2022: Proc. 39th International Conference on Machine Learning, 2022.


    2. Optimal Pricing of Information

      (α-β) Shuze Liu, Weiran Shen and Haifeng Xu

      EC 2021: Proc. 22th ACM Conference on Economics and Computation, 2021.


    3. Selling Information through Consulting

      (α-β) Yiling Chen, Haifeng Xu, Shuran Zheng

      SODA 2020: Proc. 31st ACM-SIAM Symposium on Discrete Algorithms, 2020.


  4. Incentives-aware machine learning and multi-agent learning

    5. Representative Recent Papers (see also a recent talk I gave at Princeton):

    1. Bandits Meet Mechanism Design to Combat Clickbait in Online Recommendation

      Thomas Kleine Buening, Aadirupa Saha, Christos Dimitrakakis and Haifeng Xu

      ICLR 2024: Proc. 12th International Conference on Learning Representations, 2024 (spotlight, 5%)


    2. PAC-Learning for Strategic Classification

      (α-β) Ravi Sundaram, Anil Vullikanti, Haifeng Xu, Fan Yao

      The Journal of Machine Learning Research (JMLR), 2023 (also appeared at ICML 2021 as long oral, 3%)

      See an awesome blog series illustrating this paper to public audience by Towards Data Science


    3. Multi-Agent Learning for Iterative Dominance Elimination: Formal Barriers and New Algorithms

      (α-β) Jibang Wu, Haifeng Xu and Fan Yao

      COLT 2022: Proc. 35th Annual Conference on Learning Theory, 2022.



Recent News

PHP Hits Count