BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Rule extraction from deep neural nets - Urska Matjasec (University
  of Ljubljana)
DTSTART:20191108T160000Z
DTEND:20191108T170000Z
UID:TALK134827@talks.cam.ac.uk
CONTACT:Mateja Jamnik
DESCRIPTION:Deep neural nets are a black box approach to statistical learn
 ing. The classification from the model is hard to interpret. We present an
  approach to extracting rules from a deep learning model that gets cancer 
 patients’ gene expression as input and classifies them as those likely o
 r unlikely to relapse. In this way we get some level of explainability of 
 the output from the deep learning model.
LOCATION:FC22\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR