About
A little about me.
I'm Zeru Zhu, a PhD student in Applied Mathematics at Stony Brook University. I enjoy seeing connections between different areas of mathematics and the insights they can bring.
Recently, I've been interested in AI for mathematics, spectral graph theory, representation theory, and the Langlands program. With AI, I'm especially interested in how it can help with proofs and mathematical understanding.
Insight & connections
The Langlands program is one example of the kind of mathematical connections I find interesting. The connections in my own projects are more familiar to people who work in those areas, but they illustrate two things I enjoy about mathematics.
Turning a problem into mathematics
In the epidemic-control project, allocating limited curing resources becomes a semidefinite programming problem. This is mathematical abstraction: translating a question about disease spread and resource allocation into a formulation we can analyze.
Connecting mathematical viewpoints
In algebraic-connectivity maximization, a network's connectivity is studied through the eigenvalues of its Laplacian. Here the connection is within mathematics: a question about graph structure becomes a question in linear algebra and spectral theory.
AI & mathematics
I'm interested in how AI, together with the harnesses that organize its work, can help generate proofs and enhance mathematical understanding. My work on MerLean is one way I explore this interest.
What is the right way to use AI in mathematics? How can it deepen our understanding? Could it ultimately hinder the development of mathematics? These are open questions for me.
My publication record also includes distributed optimization and other areas. This gallery records that work alongside my current interests; it is a place for what I've worked on, not just what I'm thinking about now.
Education
PhD, Applied Mathematics
Stony Brook University · New York
MS, Mathematics
New York University · New York
BS, Mathematics
Stony Brook University · New York
Toolkit
Mathematics
Optimization, graph theory, probability, dynamical systems, control, spectral methods
Learning
Reinforcement learning, multi-agent learning, transformers, GNNs, GFlowNets
Computing
Python, C++, MATLAB, Julia, PyTorch, JAX, NumPy, SciPy, Lean
Contact
Research question, possible collaboration, or just something mathematically interesting?
zeru.zhu@stonybrook.edu