Mixture Density Estimation and Hilbert Space Embedding of Measures
Bharath Sriperumdur (Gatsby Unit, University College London)
NICTA SML SEMINARDATE: 2011-10-27
TIME: 11:00:00 - 12:00:00
LOCATION: NICTA - 7 London Circuit
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
In this talk, I will consider the problem of estimating a density using a finite combination of densities from a given class, C. Unlike previous works, where Kullback-Leibler (KL) divergence is used as a notion of distance, I will consider a distance measure based on the embedding of densities into a reproducing kernel Hilbert space (RKHS) and analyze the estimation and approximation errors for an M-estimator and a greedy estimator. The advantage with the Hilbert space embedding approach is that these estimators achieve better convergence rates than those obtained with KL divergence, without making any assumptions on C, in contrast to the KL divergence approach, where the densities in C are assumed to be bounded (and away from zero) with C having a finite entropy integral.
BIO:
Bharath Sriperumdur is a research associate in Gatsby Unit, University College London and has obtained a Ph. D. in Electrical and Computer Engineering from the University of California, San Diego.
