BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Questioning Ideas in Uncertainty Estimation in Deep Learning - Guo
 xuan Xia (Imperial College London)
DTSTART:20231113T120000Z
DTEND:20231113T130000Z
UID:TALK208075@talks.cam.ac.uk
CONTACT:Simon Webster McKnight
DESCRIPTION:“Stories” in research often are simple\, clean and conveni
 ent. This talk is about some of my experiences questioning some of these s
 tories in the field of uncertainty estimation in deep learning. We will ha
 ve a look at the following questions:\n<ul>\n<li>Is ensemble diversity rea
 lly useful for uncertainty estimation?</li>\n<li>Why do we ignore incorrec
 t in-distribution predictions when evaluating Out-of-Distribution Detectio
 n?</li>\n<li>Are Deep Ensembles really too computationally costly?</li>\n<
 /ul>\n\nSpeaker bio: Guoxuan Xia is currently a PhD student at the Circuit
 s and Systems group in Imperial College London. He undertook his master’
 s project under the supervision of Prof. Mark Gales at CUED. His research 
 interests are primarily in the areas of reliability (uncertainty\, robustn
 ess) and computational efficiency (dynamic neural networks\, quantisation\
 , knowledge distillation) in deep learning.
LOCATION:Hybrid: JDB Teaching Room\, Engineering Department or Zoom: https
 ://cam-ac-uk.zoom.us/meeting/register/tZ0tfuCvrDgjG9WNY52TL0qsu7PqO484Pad5
END:VEVENT
END:VCALENDAR