Hello! I’m a third-year (as of Fall 2024) computer science Ph.D. candidate at Princeton University. Before that, I was an undergraduate student at Yale, where I studied mathematics (combined M.S./B.S.) and computer science. I’m lucky to have two advisors: Mark Braverman and Matt Weinberg.
I’ve spent much of the past two years thinking about online selection problems. I’m particuarly interested in questions about what makes certain OSPs tricky, including:
- Why can’t we seem to do better than a \(O(m)\)-competitive ratio for an intersection of \(m\) matroids, even though the best known lower bound is \(O(m^{\frac{1}{2}+o(1)})\)? Why is the gap so large?
- Why isn’t random-order enough to make matroid secretary \(O(1)\) (so far), and what properties of random-order aren’t we using?
If you have an idea for a project that you’d like to collaborate on or chat about, feel free to reach out!
Recent Publications
"*" denotes myself.
Decentralization Cheapens Corruptive Majority Attacks
*.
AFT 2023.
Optimal Rates for Bandit Nonstochastic Control
Y. Jennifer Sun, *, Elad Hazan.
NeurIPS 2023.
Contact
Email: firstname (dot) lastname (at) princeton (dot) edu, excluding spaces.