Events Calendar

Epstein Institute / ISE 651 Seminar Series

Speaker: Jorge Nocedal, David and Karen Sachs Professor of Industrial Engineering and Management Sciences, Northwestern University

Talk Title: Variance Reduction Optimization Methods for Machine Learning

Series: Epstein Institute Seminar Series

Abstract: The stochastic gradient method plays a central role in large-scale statistical learning, where vast amounts of data and high dimensional models are employed. This method is, however, difficult to tune and several variance reduction methods have been recently proposed to address this problem. After motivating these methods, I will propose a new variant that enjoys improved learning properties. Numerical results on problems arising in text classification, speech and image recognition will be presented.

Biography:
Jorge Nocedal is the David and Karen Sachs Professor of Industrial Engineering and Management Sciences at Northwestern University. He obtained a B.S. degree in physics from UNAM and a PhD in mathematical sciences from Rice University. His research interests are in optimization and its application in machine learning. He has served as the Editor in Chief of the SIAM Journal on Optimization, is a SIAM Fellow, and was awarded the 2012 George B. Danzig Prize.


Host: Daniel J. Epstein Department of Industrial and Systems Engineering

More Information: Seminar-Nocedal-2.docx

Location: Ethel Percy Andrus Gerontology Center (GER) - 206

Audiences: Everyone Is Invited

Contact: Georgia Lum