3 p.m. - 4 p.m. Location: FO 2.702
M. Vidyasagar FRS
Cecil & Ida Green Chair in Systems Biology Science
The University of Texas at Dallas
Matrix Completion and Partial Realization via Compressed Sensing
Compressed sensing refers to the problem of reconstructing large but sparse objects from a limited number of measurements. Within this broad topic, matrix completion refers to the problem of reconstructing a large but low rank matrix from measurements of just a few of its components. This is a relatively new area of research. Partial realization, in contrast, is a 50 year-old problem of completing a partially observed unit pulse response of a control system with a dynamical model of minimal order. In this talk, I will describe some work due to others (matrix completion) and some of my own on-going research (partial realization) and establish connections between the two topics.
Sponsored by the Department of Mathematical Sciences