Yichen Huang

I am a Ph.D. student in Computer Science at Harvard University under the wise guidance of Prof. Michael Mitzenmacher. Previously, I completed my undergraduate studies at the University of Oxford, where I was fortunate to work with Prof. Christian Coester and Prof. Standa Živný.

I am broadly interested in theoretical computer science, and particularly decision-making under uncertainty, including:

1. Online Algorithms: An algorithm is called online if it receives the input over time and has to make irrevokable decisions without knowing the future. How can we design algorithms that perform wellwhen comparing against the optimal algorithm in hindsight, measured in low competitive ratio (the cost of the algorithm divided by the cost of the optimum) or low regret (the cost of the algorithm less the cost of the optimum)?
2. Learning-augmented Algorithms: Inspired by the successes of machine learning, how can an algorithm utilize a prediction oracle that provides us with some hint of the problem with no guarantee of correctness? Can we design algorithms that take advantage of accurate predictions while minimizing the impact of incorrect ones?
3. Mechanism Design: Many settings in economics involve decision-making uncertainty. When the participants in the algorithm are individually trying to maximize their own utilities, what is the cost of making the algorithm truthful? How does mechanism design interact with predictions? What about online mechanism design when participants arrive sequentially?

Occasionally, I participate in empirical and experimental research, mostly by providing theoretical insights and justifications.

I enjoy talking to people with different backgrounds and distinct insights. If you are interested in a potential collaboration or would like to talk about anything, feel free to email me at [{firstname}{lastname}]@g.harvard.edu.

Research

Authors are listed in alphabetical ordering unless they are not, in which case (*) denotes equal contributions.

Theoretical Computer Science

The Mixed Birth-death/death-Birth Moran Process
David A. Brewster, Yichen Huang, Michael Mitzenmacher, Martin A. Nowak
ITCS 2026

Online and Dynamic Metric Embeddings with Monotone Recourse
Christian Coester, Yichen Huang
Manuscript

EconCS

Form vs Information: The Regimes of Data Property Rights
Yichen Huang, Zhen Sun, Xinzhen Xu, Bozhong Zheng
Manuscript

Machine Learning

Balancing Growth and Equity: Multi-Objective Optimization in Urban Renewal with Deep Reinforcement Learning
Jianghao Yu*, Yichen Huang*, Yuming Lin, Yu Zheng, Yuge Cui, Michael Mitzenmacher, Yong Li, and Li Tian
Manuscript

Efficient Task Grouping Through Samplewise Optimisation Landscape Analysis
Anshul Thakur*, Yichen Huang*, Soheila Molaei, Yujiang Wang, David A. Clifton
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025

Property Prediction of Bio-Derived Block Copolymer Thermoplastic Elastomers Using Graph Kernel Methods
Shannon R. Petersen, David Kohan Marzagão, Georgina L. Gregory, Yichen Huang, David A. Clifton, Charlotte K. Williams, Clive R. Siviour
Angewandte Chemie, 2025

Talks

Online and Dynamic Metric Embeddings with Monotone Recourse
TGINF (Harvard Student Theory Seminar) 2025
Workshop on Local Algorithms 2025

Seminars and Organizations

I currently co-organize TGINF, a theory seminar for graduate students and postdocs at Harvard.

Service

Subreviewer for ITCS 2026.

Teaching

TA for Harvard CS 2380 Optimized Democracy, Fall 2025.

Misc

Outside of academics, I enjoy playing Poker and cooking. I am recently quite into bouldering. I used to participate extensively in competitive programming, having won a gold medal at ICPC NWERC 2022 and advancing to the World Finals. I also spend some time in (Mandarin Chinese) competitive debating. I currently lead the Harvard Chinese Debate Team, and I used to be the captain of the Chinese Debate Team at Oxford, with whom I was the runner-up in the Chinese Debate World Cup.