University of Cambridge > > Inference Group > Metropolis-Coupled MCMC for Nested Sampling

Metropolis-Coupled MCMC for Nested Sampling

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact David MacKay.

In this talk, I will first introduce nested sampling, which is a recent probabilistic numerical integration algorithm.

After that, I’ll focus on the problem of sampling a point from a truncated domain which is essential to the accuracy of nested sampling. To solve this problem, I propose the technique called Metropolis-coupled MCMC (MCMCMC) which can unify a random-walk MCMC and the ellipsoid method.

The key idea is to use the ellipsoid method to create an auxiliary distribution. The auxiliary distribution then help the random-walk MCMC to converge faster. The MCMCMC framework is general and theoretically correct as the MCMC framework.

This talk is part of the Inference Group series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2023, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity