Mathematics · Imperial College London

Yifu Zhang

First-year Mathematics PhD student at Imperial College London

Numerical analysis · Randomized numerical linear algebra · Approximation theory · Operator learning

Research interest

I am interested in numerical analysis in general. I have worked in approximation theory and randomised numerical linear algebra before, and currently I am exploring their applications for areas like operator learning, Koopman operators and numerical methods for SPDEs.

Oxford MMath in Mathematics — First class.

Education

Oct 2026–present
PhD in Mathematics
Imperial College London
First year · Supervised by Nicolas Boullé, Greg Pavliotis, and Matthew Colbrook (University of Cambridge)
Oct 2022–Jun 2026
MMath in Mathematics
University of Oxford
First class · Ranked 8/85 in MMath
Graduated Jun 2022
International Baccalaureate Diploma
Keystone Academy
Beijing, China
Keystone Scholarship · Awarded 2021 and 2022

Selected work

Current research

All research

Approximation theory · Numerical analysis

Function approximation under heteroskedastic noise

HeteroChebtrunc adapts Chebyshev-based approximation to reduce effect of heteroskedastic noise.

Randomized numerical linear algebra

CUR approximation for Gram matrices

My Oxford MMath dissertation studied Gram matrices approximations using CUR.

Publication

Recent preprint

Full citation

Preprint · 2025

Efficient Function Approximation Under Heteroskedastic Noise

Yuji Nakatsukasa · Yifu Zhang

arXiv:2508.08683 · Numerical Analysis (math.NA)