University of Cambridge > > Machine Learning @ CUED > Sparse Gaussian Process in Disease Mapping

Sparse Gaussian Process in Disease Mapping

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

If you have a question about this talk, please contact Zoubin Ghahramani.

Disease mapping is a research area in spatial epidemiology, which aims to describe the overall disease distribution on a map. The aim might be, for example, to highlight areas of elevated or lowered mortality or morbidity risk. Gaussian process gives a natural prior for the log risk surface, since the spatial correlations between areas can be included in an explicit and natural way into the model via a correlation function. The drawback with using a Gaussian process is the computational burden of the covariance matrix calculations and analytically intractable model. In this talk we consider sparse approximations to Gaussian process prior to speed up the computations and approximate approaches for posterior inference. The sparse approximations are fully and partially independent conditional (FIC and PIC ) and the posterior inference is conducted with a help of Markov chain Monte Carlo methods and expectation propagation.

This talk is part of the Machine Learning @ CUED 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