Academic

Seminars Events Calendar


Filter by
Jump to
Search

Regeneration-enhanced Markov processes and application to Monte Carlo

Mr Andi Wang University of Oxford

Date:21 October 2019, Monday

Location:S16-05-96, DSAP Computer Lab 4

Time:03:00pm - 04:00pm

 

I will discuss a class of Markov processes comprising local dynamics governed by a fixed Markov process which are enriched with regenerations from a fixed distribution at a state-dependent rate. We give conditions under which such processes possess a given target distribution as their invariant measures, thus making them suitable as the basis of a new Monte Carlo sampling framework. The resulting Restore Sampler has several desirable properties: simplicity, lack of rejections, nonreversibility, regenerations and a potential coupling from the past implementation. The sampler can also be used as a recipe for introducing rejection-free moves into existing Markov Chain Monte Carlo samplers in continuous time.

Joint work with Murray Pollock, Gareth Roberts and David Steinsaltz.