Graduate Student / Postdoc Seminar

Approximating Matrix Eigenvalues by Subspace Iteration and Repeated Random Sparsification

Speaker: Jonathan Weare, NYU Courant

Location: Warren Weaver Hall 1302

Date: Friday, October 28, 2022, 1 p.m.


Traditional numerical methods for calculating matrix eigenvalues are prohibitively expensive for high-dimensional problems. I will present an iterative approach based on a non-standard subspace iteration with repeated random sparsification that can accurately estimate multiple dominant eigenvalues of matrices so large that storing even a single dense vector would be impossible.