Button to scroll to the top of the page.

Updates

Campus health and safety are our top priorities. Get the latest from UT on COVID-19.

Get help with Zoom and more.

Events Calendar

Yearly View
View as List
Monthly View
By Month
Weekly View
By Week
Daily View
Today
Search
Search
Download as iCal file
SDS Seminar Series: Yanxin Li
Friday, September 25, 2020, 02:00pm - 03:00pm
Contact : email

The Fall 2020 SDS Seminar Series continues with Yanxin Li (PhD candidate; The University of Texas at Austin, Department of Statistics and Data Sciences), on Friday, September 25th, 2020 from 2:00 p.m. to 3:00 p.m. via Zoom. Please contact stat.admin@austin.utexas.edu for the Zoom link.

SDS Logo

Title: "Latent Slice Sampling"

Abstract: In Bayesian inference, the two most popular sampling methods for constructing suitable transition kernels are the Metropolis-Hastings method and the Gibbs sampler. However, the Metropolis-Hastings method requires a good proposal distribution, otherwise the chain might become trapped in local mode or simply not move at all. To address this “sticking” problem, a Markov Chain Monte Carlo (MCMC) sampling method called latent slice sampling is proposed, to apply in a vast range of research paradigms where the Metropolis-Hastings algorithm is currently used. It is an alternative method which can replace Metropolis-Hastings method for faster convergence, better mixing of the chain, no tuning and no accept/reject component. The latent slice sampler is applied to both the discrete and continuous random variables. This talk will be an extension of my candidacy talk and mainly focuses on multivariate latent slice sampler, adaptive latent slice sampling on continuous variables, as well as theories behind the sampling algorithm.

Location: Zoom
September 25, 2020 - Yanxin Li (PhD Candidate)
The University of Texas at Austin, Department of Statistics & Data Sciences
Zoom, 2:00 to 3:00 PM
Read more:
SDS Seminar Series