Zishun's Homepage

Zishun Yu

Computer Science,
University of Illinois at Chicago
Chicago, IL 60607
E-mail: zyu32 [@] uic [DOT] edu
[LinkedIn] [Kaggle] [Google Scholar]

About Me

I am a computer science Ph.D. student at the University of Illinois Chicago (UIC), since 2021, fortunate to be advised by Prof. Xinhua Zhang. Also, I am privileged to work with Prof. Ian Kash and Prof. Lev Reyzin. My research interest spans reinforcement learning (RL) and its application in large language models (LLMs).

Prior to that, I received my master's degree in IEOR from UIC, under the advisory of Prof. Mengqi Hu, and my bachelor's degree from Huazhong University of Science and Technology (HUST).

News

  • 2024-03: Awarded travel support from ICLR 2024, see you in Vienna. 🇦🇹
  • 2024-02: Presented at ALT, San Diego, California.
  • 2024-01: One paper accepted at ICLR 2024 as spotlight presentation.
  • 2023-12: One paper accepted at ALT 2024.
  • 2023-11: One paper accepted at FMDM workshop @ NeurIPS.
  • 2023-09: Continuing my internship with TikTok @ Chicago as a research scientist intern.
  • 2023-08: Presented at ICML, Honolulu, Hawaii. 🏝
  • 2023-05: Joining TikTok @ Bellevue as a research scientist intern.
  • 2023-04: One paper accepted at ICML 2023.
  • 2023-01: Received Kaggle Competition Master title.
  • 2023-01: Won a gold medal 🥇 for Kaggle annual optimization competition 2022.
  • 2022-12: Presented at NeurIPS, New Orleans, Louisiana.
  • 2022-09: One paper accepted at NeurIPS 2022.
  • 2022-05: One paper accepted at UAI 2022.
  • 2022-01: Won a silver medal 🥈 for Kaggle annual optimization competition 2021.
  • 2021-09: One paper accepted at IEEE Trans. ITS.
  • 2021-08: Starting as a Ph.D. student in CS at UIC.
  • 2021-05: Received NSF TRIPODS graduate fellowship (summer 2021).
  • 2021-05: Received M.Sc. in IEOR from UIC.
  • 2021-03: Won a bronze medal 🥉 for Kaggle RANZCR CLiP competition.
  • 2021-01: Won a silver medal 🥈 for Kaggle annual optimization competition 2020.
  • 2018-09: Presented at INFORMS annual meeting, Phoenix, Arizona

Publications/Manuscripts

* equal contribution/main student author
in alphabetical order

  • [1] \(\mathcal{B}\)-Coder: On Value-Based Deep Reinforcement Learning for Program Synthesis. [PDF]
    Zishun Yu, Yunzhe Tao, Liyu Chen, Tao Sun and Hongxia Yang
    ICLR [Spotlight] (International Conference on Learning Representations), 2024
    FMDM@NeurIPS (Foundation Models for Decision Making Workshop), 2023

  • [2] Slowly Changing Adversarial Bandit Algorithms are Efficient for Discounted MDPs. [PDF]
    Zishun Yu*, with Ian A. Kash and Lev Reyzin
    ALT (International Conference on Algorithmic Learning Theory), 2024

  • [3] Actor-Critic Alignment for Offline-to-Online Reinforcement Learning. [PDF][Code]
    Zishun Yu and Xinhua Zhang
    ICML (International Conference on Machine Learning), 2023

  • [4] Certifying Robust Graph Classification under Orthogonal Gromov-Wasserstein Threats. [PDF][Code]
    Hongwei Jin, Zishun Yu and Xinhua Zhang
    NeurIPS (Neural Information Processing Systems), 2022

  • [5] Orthogonal Gromov-Wasserstein Discrepancy with Efficient Lower Bound. [PDF][Code]
    Hongwei Jin, Zishun Yu and Xinhua Zhang
    UAI (Uncertainty in Artificial Intelligence), 2022

  • [6] Deep Reinforcement Learning with Graph Representation for Vehicle Repositioning. [PDF]
    Zishun Yu and Mengqi Hu
    IEEE-ITS (IEEE Transactions on Intelligent Transportation Systems), 2021

Professional Experience

  • TikTok/ByteDance Inc. - Applied Machine Learning team (Remote) @ Chicago IL
    Research Scientist Intern, Aug 2023 - Sept 2023
    Large Language Models (LLMs), RL fine-tuning, Computer Program Synthesis

  • TikTok/ByteDance Inc. - Applied Machine Learning team @ Bellevue WA
    Research Scientist Intern, May 2023 - Aug 2023
    Large Language Models (LLMs), RL fine-tuning, Computer Program Synthesis

Awards

  • MITACS Globalink Undergraduate Research Internship
    York University, Canada

Professional Services

Conference Reviewer/Program Committee Member:

  • Conference on Neural Information Processing Systems (NeurIPS) , 2022 - 2023

  • International Conference on Machine Learning (ICML) , 2023 - 2024

  • International Conference on Learning Representations (ICLR) , 2023 - 2024

  • International Conference on Artificial Intelligence and Statistics (AISTATS) , 2023 - 2024

  • Conference on Uncertainty in Artificial Intelligence (UAI) , 2023 - 2024

  • Foundation Models for Decision Making @ NeurIPS 2023

Journal Reviewer:

  • Transactions on Machine Learning Research (TMLR) , 2023 -

  • IEEE Computational Intelligence Magazine (CIM) , 2023 -

Competitions

Kaggle competition Master¹. (top 1% of all Kaggle users)

  • [Kaggle] Santa's Annual Optimization Challenge 2022 [Link]
    Gold medal          (rank 8/875)
    Robotics, Planning, Optimization, Combinatorics, TSP

  • [Kaggle] Santa's Annual Optimization Challenge 2021 [Link]
    Silver medal        (top 4%)
    Optimization, Combinatorics, Superpermutation, Coloured TSP

  • [Kaggle] Santa's Annual Optimization Challenge 2019 [Link]
    Silver medal        (top 5%)
    Optimization, Combinatorics, Assignment Problem, Mixed Integer Programing

  • [Kaggle] Catheter and Line Position Challenge [Link]
    Bronze medal     (top 8%)
    Classification, Computer Vision, Medical Image

(¹Kaggle official ranking system.)