Yuxiao Wen

temp_profile_pic.png

Yuxiao Wen (top)

I am a Ph.D. candidate in Computer Science in Courant Institute of Mathematical Sciences at New York University. I am fortunate to be advised by Prof. Yanjun Han and Prof. Zhengyuan Zhou. My research primarily focuses on reinforcement learning and decision-making, both on theory and applications. In particular, it spans post training, contextual bandits, and counterfactual estimation.

In 2021-23, I have been working on Scientific Machine Learning with Prof. Benjamin Peherstorfer. Prior to that, I received my B.A.s in Math and in CS at New York University and graduated with latin honor and department award.

More about me: I (fortunately) have the most lovely cat in the world. I love anything with matcha. My favorite anime/manga are (alphabetical order) Ajin, Alice in Borderland, Fire Punch, Jojo 2-4, The Disastrous Life of Saiki K., and tons of others.

Links to LinkedIn and Google Scholar.

My CV.

I'm graduating in 2027 and seeking full-time roles!

You can contact me via email: yuxiaowen@nyu.edu

Interests

  • Reinforcement learning

  • Contextual bandits

  • Online decision-making

Education

  • Ph.D. in CS, New York University, 2021-present

  • B.A. in CS, New York University, 2017-2021

  • B.A. in Honors Math, New York University, 2017-2021

Selected publications

  1. Optimal Arm Elimination Algorithms for Combinatorial Bandits
    Yuxiao Wen ,  Yanjun Han ,  and  Zhengyuan Zhou
    Conference on Artificial Intelligence and Statistics (AISTATS), 2026
  2. Joint Value Estimation and Bidding in Repeated First-Price Auctions
    Yuxiao Wen ,  Yanjun Han ,  and  Zhengyuan Zhou
    2026
    POMS-HK 2026 Best Student Paper Finalist
    POMS-China 2026 Best Student Paper Third Prize

Ongoing works

  1. Trustworthy On-policy Distillations under Operational Constraints
    with Yifei Pang, Yanjun Han, and Zhun Deng
    Working draft
  2. Learning to Activate and Scale: Price-based Semi-bandit Algorithms for Resource Allocation
    with Jiawen Yang, Caihua Chen, Xingyu Lu, Wei Lu, and Zhengyuan Zhou
    Working draft
  3. Constrained Gradient Equilibrium: Approachability under Locally Constrained Adversaries
    with Kaicheng Zhang, Lujing Zhang, Yifei Pang, Zhun Deng, and Yanjun Han
    submitted to AISTATS 2027
  4. Functional Critics Are Essential for Actor-Critic: From Off-Policy Stability to Efficient Exploration
    with Qinxuan Bai, Yuxuan Han, Wei Xu, and Zhengyuan Zhou
    submitted to ICLR 2027
  5. Online Inventory Control with Context-Aware Demand Distributions
    with Jingkai Huang, Huawei Zhou, and Zhengyuan Zhou
    reject&resubmit at M&SOM

Teaching assistant