Gilbert Strang


Speaker: Gilbert Strang
Title: Singular Values in Linear Algebra and Data Science
Affiliation: MIT
Date: Monday, March 20, 2017
Place and Time: Room 101, Love Building, 3:35-4:30 pm
Refreshments: Room 204, Love Building, 3:00 pm

Abstract. Singular values have moved into a central position in applied linear algebra. The SVD is the method of choice to identify the important part in a matrix of data. This talk will try to organize the key ideas -- for the theory as well as applications. A valuable property is to identify the important part of a matrix as a sum of rank-one pieces --- ordered by the size of the singular values. Biologists studying human evolution have moved up to data sizes of 10^5 -- a revolution for understanding.