吴越 / Yue WU. *
PhD student @ USC
Third-year PhD in CS @ Viterbi USC. Interested in algorithmic game theory & learning theory.
*: The correct order of my name is WU Yue, and 吴 / Wu is my surname. ↩
Hi, my name is
About Me
My name in Mandarin is pronounced as [u˧˥.ɥœ˥˩] in IPA. As an approximation, the first syllable (surname) is how you would pronounce Wu in English, and the second syllable (given name) is similar to “suer” in French without the leading s, or “üä” in German.
I am a second-year PhD student at the Thomas Lord Department of Computer Science at USC Viterbi, currently working in the lab of Prof. Haipeng Luo in the Theory Group. My current research topics including online learning in games, bandit learning, and algorithmic game theory in general.
Aside from research, I also maintain a PGP Keyserver, and run a DN42 AS AS4242423961. I broadly enjoy video games, especially puzzlers and speedrunning. I also enjoy the animation franchise My Little Pony: Friendship is Magic and its prolific fan music scene.
Publication
Working Papers
Most papers use alphabetical author order following convention in TCS and math.
-
“Accelerating min-max optimization via power-law stepsizes.”
Yue Wu, Weiqiang Zheng, Yang Cai, Haipeng Luo, 2026.
Available on [arXiv].
Conference Publications
Most papers use alphabetical author order following convention in TCS and math.
-
“Adversarial learning in games with bandit feedback: Logarithmic pure strategy maximin regret.”
Shinji Ito, Haipeng Luo, Arnab Maiti, Taira Tsuchiya, Yue Wu, 2026.
To appear in COLT 2026: the 39th Conference on Learning Theory.
Available on [arXiv]. -
“Instance-dependent regret bounds for learning two-player zero-sum games with bandit feedback.”
Shinji Ito, Haipeng Luo, Taira Tsuchiya, Yue Wu, 2025.
COLT 2025: the 38th Conference on Learning Theory.
Available on [arXiv] [PMLR]. -
“Sybil-proof diffusion auction in social networks.”
Hongyin Chen, Xiaotie Deng, Ying Wang, Yue Wu, Dengji Zhao, 2023.
AAMAS 2023: the 22nd International Conference on Autonomous Agents and Multiagent Systems.
Available on [arXiv] [ACM DL].
Education / Experience
I worked as a student intern at RIKEN AIP, Tokyo, Japan.
I did research on online learning in games, specifically on bandit-feedback learning in two-player zero-sum games.
I work with an amazing group of researchers at the Sequential Decision Making Team,
including Prof. Shinji Ito
and Prof. Taira Tsuchiya.
I am a third-year PhD student at USC Viterbi, working with Prof. Haipeng Luo in the Theory Group.
Teaching
- I TAed for the course Machine Learning for Data Science by Dr. Mohammad Reza Rajati in the summer of 2023 and the spring of 2026.
- I TAed for the course Machine Learning by Prof. Haipeng Luo in the fall of 2025.
I worked as an RA intern at Ant Group, Hangzhou, China. Working with a talented group with backgrounds in optimization, finance and operatics, I did research on how large recommendation platforms can effectively steer the market of content creators and consumers.
Bachelor of Science (Major: Data Science and Big Data Technology)
-
Peking University
Sept 2019 - June 2023
I graduated from Peking University in the spring of 2023 with a major in Data Science and Big Data Technology.
Achievements
- Graduated magna cum laude in the spring of 2023.
- Merit Student for three academic years in a row (2019-2021) for demonstrating outstanding performance.
- Baosteel Scholarship (2020)
- The Third Prize of Peking University Scholarship (2021)
Experiences
- Participated in the Collegiate Programming Contests, and won a Gold Medal in the 2019 ICPC Asia Yinchuan Regional Contest (15th place out of 242 teams).
- Participated in research at the Center on Frontiers of Computing Studies under the supervision of Prof. Xiaotie Deng.
- TAed for the course Data Structures and Algorithms in the fall of 2022 by Prof. Qun HUANG.
Projects
Typinox
A Python library that supports enhanced run-time typechecking for
jaxtyping annotated equinox Modules.
Lower bound of deterministic Zipfian top-k problem
A complexity analysis for a specific data stream algorithm problem.
Course project for Algorithm Design and Analysis (Honor Track) by Dr. Tong YANG.
WTSYC
A toy compiler for a C subset, written in C++ (150KB code).
Project for our Compiler course in fall 2021.
Superliminal Auto-splitters
Superliminal: a 2019 puzzle-platformer by Pillow Castle. Speedrunning: trying to beat a video game as fast as possible. Auto-splitter: a small script of code that detects the start, stages and end of a video game, usually to time speedruns.
I’m the maintainer of the auto-splitters for Superliminal since 2021. The development requires reverse-engineering the game.
Get in Touch
My inbox is always open. Whether you have a question or just want to say hi, I’ll get back to you as soon as possible!