Research
My research interests sit at the intersection of algorithms, complexity, and mathematical computation. I am especially drawn to problems where the right abstraction simplifies a seemingly difficult question.
This work studies a variant of 3SUM in which preprocessing is of more benefit than it looks like. The paper presents a cleaner algorithmic route and improved complexity bounds, building around FFT, hashing, and the idea that preprocessing can be used as a serious structural advantage rather than a minor convenience.
Open on arXiv Journal LinkProjects
I worked on improving bounds for a preprocessing-heavy 3SUM variant, reducing the time complexity from \( O(N^{11/6}) \) to \( O(N^{1.5} \log N) \). The project also led to questions about deterministic methods and unknown-set configurations in the preprocessing phase.
This project studied how Continuous Learning with Errors can provide stronger hardness assumptions in clustering, especially around k-means. The work was motivated by the broader question of how cryptographic hardness can reshape learning theory.