Research
Research
Seminars
Events Calendar
Seminars
Randomized Newton methods for large least-squares problems
3:30 pm
POB 6.304
We discuss randomized Newton and randomized quasi-Newton approaches to efficiently solve large linear least-squares problems, where the very large data sets present a significant computational burden (e.g., the size may exceed computer memory or data are collected in real-time). In our proposed framework, stochasticity is introduced as a means to overcome computational limitations, and probability distributions that can exploit structure and/or sparsity are considered. Our results show, in particular, that randomized Newton iterates, in contrast to randomized quasi-Newton iterates, may not converge to the desired least-squares solution. Numerical examples, including an example from extreme learning machines, demonstrate the potential applications of these methods.
Sign Up for Seminar Announcements
To sign up for our weekly seminar announcements, send an email to sympa@utlists.utexas.edu with the subject line: Subscribe ase-em-seminars.