Gavin (Jialun) Zhang

Phone: (909) 575-7644

About Me

Hello! My name is Gavin. I'm a research scientist on the Kernels and Optimizations Team at Meta Superintelligence Labs . I was previously part of the AI and Systems Co-Design team. My research focuses on training algorithms for large language models and large-scale recommendation systems. Currently I am working on the Distributed Shampoo optimizer.

I finished my PhD in 2024 from the ECE department at the University of Illinois at Urbana-Champaign, where I worked on preconditioned gradient methods for low-rank matrix optimization. My advisor was Richard Y. Zhang.

Previously, I received a bachelor's degree in mathematics from Harvey Mudd College, and a master's degree in computational mathematics from Stanford.

Selected Publications

Preconditioned Gradient Descent for Overparameterized Nonconvex Burer-Monteiro Factorization with Global Optimality Certification

Journal of Machine Learning Research (JMLR 2023)

Gavin Zhang, Salar Fattahi, Richard Y. Zhang

Fast and Accurate Estimation of Low-Rank Matrices from Noisy Measurements via Preconditioned Non-Convex Gradient Descent

Proceedings of The 27th International Conference on Artificial Intelligence and Statistics (AISTATS 2024)

Gavin Zhang, Hong-Ming Chiu, Richard Y. Zhang

How Many Samples is a Good Initial Point Worth in Low-rank Matrix Recovery?

Advances in Neural Information Processing Systems (NeurIPS 2020, Spotlight Presentation)

Gavin Zhang, Richard Y. Zhang