BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Bayesian Approach to Machine Learning - Fergus Simpson -- PROWLE
 R.io
DTSTART:20191121T130000Z
DTEND:20191121T143000Z
UID:TALK130552@talks.cam.ac.uk
CONTACT:James Fergusson
DESCRIPTION:Conventional approaches to machine learning can suffer from a 
 wide range of issues such as overfitting\, poorly calibrated uncertainties
 \, and difficulty in explaining their outputs.  I will outline various ste
 ps which have been taken towards resolving these issues\, by adopting a pr
 obabilistic framework.  This includes some of the latest research from PRO
 WLER.io\, where we apply Bayesian inference to a wide range of machine lea
 rning problems. 
LOCATION:Kavli Large Meeting Room\, Kavli Building
END:VEVENT
END:VCALENDAR