Wednesday, January 15, 2014 9:30 am
-
9:30 am
EST (GMT -05:00)
Data Fitting and Optimization With Randomized Sampling
Speaker: | Michael Friedlander |
---|---|
Affiliation: | University of British Columbia |
Room: | Mathematics and Computer Building (MC) 5168 |
Abstract:
For many structured data-fitting applications, stochastic gradient methods offer inexpensive iterations by sampling only subsets of the data. They make great progress initially, but eventually stall. Full-gradient methods, in contrast, often achieve steady convergence, but may be prohibitively expensive for large problems. Applications in machine learning and robust seismic inversion motivate us to develop a sampling method that exhibits the benefits of both stochastic and full-gradient methods.