[Ml-stat-talks] Thurs: Andriy Norets on conditional density estimation
mimno at CS.Princeton.EDU
Mon Oct 10 17:09:33 EDT 2011
For this week's Machine Learning lunchtime talk we welcome a guest
from our Nobel-prize-winning Economics department, Andriy Norets.
Note that the talk will be on Thursday, but at the usual time (12:30)
in the usual place (CS 402).
Title: Posterior consistency in conditional density estimation by
covariate dependent mixtures
(with Justinas Pelenis)
This paper considers Bayesian nonparametric estimation of conditional
densities by countable mixtures of location-scale densities with
covariate dependent mixing probabilities. The mixing probabilities
are modeled in two ways. First, we consider finite covariate
dependent mixture models, in which the mixing probabilities are
proportional to a product of a constant and a kernel and a prior on
the number of mixture components is specified. Second, we consider
kernel stick-breaking processes for modeling the mixing probabilities.
We show that the posterior in these two models is weakly and strongly
consistent for a large class of data generating processes.
A related paper can be found at http://www.princeton.edu/~anorets/consmixreg.pdf
More information about the Ml-stat-talks