BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian nonparametric dynamic-clustering and genetic imputation -
  Lloyd Elliott (Gatsby Unit\, UCL / Oxford)
DTSTART:20140304T110000Z
DTEND:20140304T120000Z
UID:TALK50977@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:I will describe new approaches to dynamic-clustering based on 
 Bayesian nonparametric (BNP) hidden Markov models (HMMs). I will apply the
 se approaches to genotype imputation problems and illustrate the practical
  benefits of BNP. Genetic similarity within a population is a function of 
 chromosome position and dynamic-clustering based on parametric HMMs are po
 pular models of genetic structure. BNP priors are well suited as extension
 s of\, or competitors to\, these HMMs because many aspects of genetic proc
 esses (such as allele sampling) arise naturally from BNP models. In additi
 on\, BNP priors provide several practical benefits over parametric HMMs. F
 irst\, by defining probability distributions on the set of partitions\, BN
 P priors avoid label switching problems. Second\, costly model selection a
 nd ad-hoc methods to determine the number of latent clusters are also avoi
 ded. Finally\, the flexibility of BNP often provides state-of-the-art impu
 tation accuracy. I will conclude with directions of future work including 
 the abstraction of the auxiliary Gibbs scheme (which I derived for inferen
 ce) to probabilistic programming for BNP models.
LOCATION:Engineering Department\, CBL Room BE-438
END:VEVENT
END:VCALENDAR