Michael Goldstein (University of Durham)

Geometric Bayes
Wednesday 22 March 2017 at 13.00, JCMB 5326

Abstract

The Bayesian approach to uncertainty quantification is popular, powerful and successful. We will describe an alternative geometric formulation for this approach, based around viewing expectation, rather than probability as the primitive quantity for uncertainty quantification. The geometric view greatly simplifies the formulation and solution of many large and complex problems in uncertainty quantification, and leads to several natural and powerful extensions. The approach will be illustrated by an example in uncertainty quantification for global climate simulators.

Seminars by year

Current 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 2000 1999 1998 1997 1996