Multiplicative Algorithms: Why Do They Converge and How to Make Them Faster

RAND Statistics Seminar Series

Multiplicative Algorithms: Why Do They Converge and How to Make Them Faster

Dr. Yaming Yu—UC Irvine

Thursday, October 13th, 2011
10:30 a.m. – 12:00 p.m. PT / 1:30pm – 3:00pm ET
Conference Room 3312
RAND Corporation, Santa Monica, CA

Please contact Fabiola Lopez if you would like to attend this seminar.

Abstract

Multiplicative algorithms are simple optimization schemes that iteratively adjust the coordinates of an input probability vector by multiplicative factors so as to increase a suitable objective function. Examples include the EM algorithm for maximum likelihood estimation of mixture proportions, multiplicative algorithms for computing approximate optimal designs, and the Arimoto-Blahut algorithm for calculating channel capacities in Shannon theory. By exploiting the connections among these seemingly separate problems we derive general conditions that ensure monotonic convergence for multiplicative algorithms in optimal designs, and construct hybrid algorithms that converge faster but maintain the simplicity and stability.

Speaker Bio

Dr. Yaming Yu is an associate professor in the Department of Statistics at the University of California, Irvine. He received his Ph.D. from the Department of Statistics at Harvard University and his B.S. in Mathematics from Beijing University. His research interests include statistical computing, information theory, Bayesian analysis and applied probability.

Attending a Seminar

Other Locations/Times:
Washington, D.C. Conf. Rm. 4132: 1:30 p.m. ET
Pittsburgh Conf. Rm. 6202: 1:30 p.m. ET

RAND visitors are welcome to attend and must RSVP at least one day prior to the seminar. To ensure your attendance please contact Fabiola Lopez at flopez@rand.org with your name, company (or university) affiliation, and national citizenship (for security purposes).

For parking and directions to RAND's Santa Monica office, please see: http://www.rand.org/about/locations/santa-monica.html.

For parking and directions to RAND's Pittsburgh office, please see: http://www.rand.org/about/locations/pittsburgh.html.

For parking and directions to RAND's Washington office, please see: http://www.rand.org/about/locations/washington.html.

For further information and to be added to the mailing list contact Fabiola Lopez at flopez@rand.org.