Fan Yao

Fan Yao

Ph.D. student at CS@UVa

University of Virginia

Biography

I am a final year Ph.D. student in CS at University of Virginia and I will be on academia job market this Fall! My advisors are Prof. Haifeng Xu (now at University of Chicago) and Prof. Hongning Wang (now at Tsinghua University). Prior to my Ph.D. journey, I worked as an algorithm engineer in Alibaba Group, building recommender systems for Taobao, the world’s largest eCommerce platform. Before that, I obtained my Bachelor’s and Master’s degrees in computational mathematics from Peking University. My research interests include algorithmic game theory, machine learning, and their applications in recommendation ecosystems and online content markets.

Download my resumé.

Interests
  • Online learning
  • Multi-agent learning
  • Algorithmic game theory
  • Information Retrieval
Education

  • Ph.D. in Computer Science, 2025

    University of Virginia

  • MSc. in Computational Mathematics, 2016

    Peking University

  • BSc. in Computational Mathematics, 2013

    Peking University

Working Papers

indicates equal contribution, authors in alphabetical order.
The Complexity of Tullock Contests
Yu He, Fan Yao, Yang Yu, Xiaoyun Qiu, Minming Li, Haifeng Xu
Last updated on Dec 9, 2024
PDF
Single-Agent Poisoning Attacks Suffice to Ruin Multi-Agent Learning
Fan Yao, Yuwei Cheng, Ermin Wei, Haifeng Xu
Last updated on Oct 31, 2024
Strategic Filtering for Content Moderation: Free Speech or Free of Distortion?
Saba Ahmadi, Avrim Blum, Haifeng Xu, Fan Yao
Last updated on Oct 31, 2024
Learning from Imperfect Human Feedback: a Tale from Corruption-Robust Dueling
Yuwei Cheng, Fan Yao, Xuefeng Liu, Haifeng Xu
Last updated on Oct 31, 2024
PDF Cite Source Document

Publications

indicates equal contribution, authors in alphabetical order.
Fan Yao, Yiming Liao, Jingzhou Liu, Shaoliang Nie, Qifan Wang, Haifeng Xu, Hongning Wang (2024). Unveiling User Satisfaction and Creator Productivity Trade-Offs in Recommendation Platforms.
Neurips, 2024.

PDF Video Source Document

Fan Yao, Chuanhao Li, Denis Nekipelov, Hongning Wang, Haifeng Xu (2024). Human vs Generative AI in Content Creation Competition: Symbiosis or Conflict?.
ICML, 2024
Oral presented at Econometric Society Interdisciplinary Frontiers (ESIF) conference on Economics and AI+ML.

PDF Source Document

Fan Yao, Yiming Liao, Mingzhe Wu, Chuanhao Li, Yan Zhu, James Yang, Jingzhou Liu, Qifan Wang, Haifeng Xu, Hongning Wang (2024). User Welfare Optimization in Recommender Systems with Competing Content Creators.
KDD, 2024.

PDF Source Document

Fan Yao, Chuanhao Li, Karthik Abinav Sankararaman, Yiming Liao, Yan Zhu, Qifan Wang, Hongning Wang, Haifeng Xu (2023). Rethinking Incentives in Recommender Systems: Are Monotone Rewards Always Beneficial?.
Neurips, 2023.

PDF Source Document

Fan Yao, Chuanhao Li, Denis Nekipelov, Hongning Wang, Haifeng Xu (2023). How Bad is Top- Recommendation under Competing Content Creators?.
ICML, 2023 (oral presentation, 2.4%).

PDF Video Source Document

Fan Yao, Chuanhao Li, Denis Nekepalov, Hongning Wang, Haifeng Xu (2022). Learning from a Learning User for Optimal Recommendations.
ICML, 2022
Selected as spotlight presentation at ICML, 2023, Workshop on Interactive Learning with Implicit Human Feedback.

PDF Cite Source Document

Jibang Wu, Haifeng Xu, Fan Yao (2022). Multi-Agent Learning for Iterative Dominance Elimination: Formal Barriers and New Algorithms.
COLT, 2022
Extended version under major revision at JMLR.

PDF Cite Source Document

Fan Yao, Chuanhao Li, Denis Nekepalov, Hongning Wang, Haifeng Xu (2022). Learning the Optimal Recommendation from Explorative Users.
AAAI, 2022.

PDF Cite Source Document

Fan Yao, Renqin Cai, Hongning Wang (2022). Reversible Action Design for Combinatorial Optimization with Reinforcement Learning.
AAAI-ML4OR, 2022.

PDF Cite Source Document

Ravi Sundaram, Anil Vullikanti, Haifeng Xu, Fan Yao (2021). PAC-Learning for Strategic Classification.
The Journal of Machine Learning Research (JMLR)
Shorter version at ICML, 2021 (oral presentation, 3%).

PDF Cite Poster Video Source Document

Experience

 
 
 
 
 
University of Virginia
Graduate Research Assistant
University of Virginia
Aug 2019 – Present Charlottesville, US
Ph.D. candidate in CS department
 
 
 
 
 
University of Chicago
Visiting Student
University of Chicago
May 2023 – Present Chicago, IL
Multi-agent and economic modeling for online content market.
 
 
 
 
 
Meta Research
Student Researcher
Meta Research
Dec 2023 – Jul 2024 Menlo Park, CA
  • Host: Dr. Qifan Wang
  • Deploy mechanism design solutions for improving user engagement on Instagram Reels.
 
 
 
 
 
Google Research
Student Researcher
Google Research
May 2022 – Sep 2022 Mountain View, CA
  • Manager: Dr. Chih-wei Hsu, Dr. Craig Boutilier
  • Bayesian preference elicitation in interactive recommender systems using Concept Activation Vectors.
 
 
 
 
 
ByteDance (TikTok) AML Lab
Research Intern
ByteDance (TikTok) AML Lab
May 2021 – Aug 2021 Seattle, WA (remote)
  • Manager: Dr. Chong Wang, Dr. Taiqing Wang
  • Enhance the recommendation diversity and mitigate the echo chamber effect via collaborative Thompson sampling approach and gradient-based Determinantal Point Processes.
 
 
 
 
 
Alibaba Group, Taobao Recommendation Team
Algorithm Engineer
Alibaba Group, Taobao Recommendation Team
Jul 2017 – Aug 2019 Hangzhou, China
Design and maintain content recommendation system for Taobao main page, focusing on deep-learning based match and ranking solution.

Featured Publications

Unveiling User Satisfaction and Creator Productivity Trade-Offs in Recommendation Platforms
It is widely observed that recommendation algorithms influence creators' productivity on content recommendation platforms. We reveal a fundamental trade-off between user satisfaction and content production volume: a relevance-driven recommendation policy with low exploration strength boosts short-term user satisfaction but undermines creators' productivity, whereas a more aggressive exploration policy can increase total content creation volume.
Fan Yao
Last updated on Nov 12, 2024 1 min read poster
Unveiling User Satisfaction and Creator Productivity Trade-Offs in Recommendation Platforms
Human vs Generative AI in Content Creation Competition: Symbiosis or Conflict?
We introduce a competition model generalized from the Tullock contest to analyze the tension between human creators and GenAI. Our theory and simulations suggest that despite challenges, a stable equilibrium between human and AI-generated content is possible.
Fan Yao
Last updated on Oct 31, 2024 1 min read poster
Human vs Generative AI in Content Creation Competition: Symbiosis or Conflict?
User Welfare Optimization in Recommender Systems with Competing Content Creators
Our mechanism is deployed on Instagram Reels and the A/B test showed promising results!
Fan Yao
Last updated on Oct 31, 2024 1 min read poster
User Welfare Optimization in Recommender Systems with Competing Content Creators
Rethinking Incentives in Recommender Systems: Are Monotone Rewards Always Beneficial?
We establish a social welfare maximization impossibility theorem in content creation competition game and propose theoretically optimal mechanisms for incentivizing creators in online content platforms.
Fan Yao
Last updated on Oct 31, 2024 1 min read poster
Rethinking Incentives in Recommender Systems: Are Monotone Rewards Always Beneficial?
How Bad is Top-K Recommendation under Competing Content Creators?
We measure the user welfare loss under the content supply equilibria induced by relevance-driven recommender systems.
Fan Yao
Last updated on Oct 31, 2024 1 min read poster
How Bad is Top-K Recommendation under Competing Content Creators?
See all posts

Contact