AUEB STATS SEMINARS 8/6/2017: Scalable Approximation Algorithms for Bayesian Variable Selection by Feng Liang

AUEB STATISTICS SEMINAR SERIES JUNE 2017



Feng Liang
Associate Professor, Department of Statistics
University of Illinois at Urbana-Champaign

Scalable Approximation Algorithms for Bayesian Variable Selection

THURSDAY 7/6/2017
12:15

ROOM 607, 6th FLOOR,
POSTGRADUATE STUDIES BUILDING
(EVELPIDON & LEFKADOS)

ABSTRACT

There has been an intense development on the estimation of a sparse regression/classification model in statistics, machine learning and related fields. In this talk, we focus on the Bayesian approach to this problem, where sparsity is incorporated by the so-called spike-and-slab prior on the coefficients. Instead of replying on MCMC for posterior inference, we have developed scalable algorithms that approximate the posterior distribution and can process data batch by batch without loading all the data into memory. Asymptotic analysis of our approach, as well as empirical evaluation, will be presented.

Facebook event: https://www.facebook.com/events/185933175266705/