BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian deep learning - Yarin Gal (Oxford)
DTSTART:20180313T140000Z
DTEND:20180313T150000Z
UID:TALK99742@talks.cam.ac.uk
CONTACT:Damon Wischik
DESCRIPTION:Bayesian models are rooted in Bayesian statistics and easily b
 enefit from\nthe vast literature in the field. In contrast\, deep learning
  lacks a solid\nmathematical grounding. Instead\, empirical developments i
 n deep learning are\noften justified by metaphors\, evading the unexplaine
 d principles at play.\nThese two fields are perceived as fairly antipodal 
 to each other in their\nrespective communities. It is perhaps astonishing 
 then that most modern deep\nlearning models can be cast as performing appr
 oximate inference in a\nBayesian setting. The implications of this are pro
 found: we can use the rich\nBayesian statistics literature with deep learn
 ing models\, explain away many\nof the curiosities with this technique\, c
 ombine results from deep learning\ninto Bayesian modeling\, and much more.
 \n\nIn this talk I will review a new theory linking Bayesian modeling and 
 deep\nlearning and demonstrate the practical impact of the framework with 
 a range\nof real-world applications. I will also explore open problems for
  future\nresearch—problems that stand at the forefront of this new and e
 xciting\nfield.\n
LOCATION:Centre for Mathematical Sciences\, MR4
END:VEVENT
END:VCALENDAR