Hi, I’m Aaron.

I am a PhD student in the Theory of Computation Lab at the University of Michigan, Ann Arbor, where I am advised by Michał Dereziński. Previously, I studied computer science and mathematics at the University of Massachusetts Amherst, where I was advised by Cameron Musco.

I work on the design and analysis of fast, scalable algorithms for modern data science and machine learning. My current research focuses on sublinear-time methods for large-scale linear algebraic computations, including low-rank approximation. My work draws on tools from probability, theoretical computer science, and optimization.

I am supported by the NSF Graduate Research Fellowship.

interests randomized algorithms · numerical linear algebra · ML theory

Email Scholar GitHub

CURRENTLY THINKING ABOUT

Suppose we have entry-wise access to a positive semidefinite matrix

A\in\mathbb{R}^{n\times n}

. Is it true that

\Omega(n/\epsilon^2)

entry accesses are required in the worst case to approximate the squared Frobenius norm of

A

up to

(1\pm \epsilon)

error, i.e., output

\widehat F

satisfying

(1-\epsilon) \| A \|^2_F \leq \widehat F \leq (1+\epsilon) \| A \|^2_F?

01 — Publications

Selected publications

[T26]

Fast Length-Squared Sampling for Positive-Semidefinite Matrix Approximation

Aaron Tian

Preprintundergraduate thesisPDF

Full list of publications →

02 — Writing

News & notes

2026.04.15news

I committed to the PhD program at UMich CSE

2026.04.13news

I received the NSF Graduate Research Fellowship

All writing & updates →

03 — Contact

Get in touch

Feel free to reach out if you would like to chat! The fastest way to reach me is email.

aatian [at] umich [dot] edu

ELSEWHERE

Google Scholar →GitHub →