BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Composite Feature Selection using Deep Ensembles - Alex Norcliffe 
 - University of Cambridge
DTSTART:20230214T130000Z
DTEND:20230214T140000Z
UID:TALK195259@talks.cam.ac.uk
CONTACT:Mateja Jamnik
DESCRIPTION:In many real world problems\, features do not act alone but in
  combination with each other. For example\, in genomics\, diseases might n
 ot be caused by any single mutation but require the presence of multiple m
 utations. Prior work on feature selection either seeks to identify individ
 ual features or can only determine relevant groups from a predefined set. 
 We investigate the problem of discovering groups of predictive features wi
 thout predefined grouping. To do so\, we define predictive groups in terms
  of linear and non-linear interactions between features. We introduce a no
 vel deep learning architecture that uses an ensemble of feature selection 
 models to find predictive groups\, without requiring candidate groups to b
 e provided. The selected groups are sparse and exhibit minimum overlap. Fu
 rthermore\, we propose a new metric to measure similarity between discover
 ed groups and the ground truth. We demonstrate the utility of our model on
  multiple synthetic tasks and semi-synthetic chemistry datasets\, where th
 e ground truth structure is known\, as well as an image dataset and a real
 -world cancer dataset.\n\n"You can also join us on Zoom":https://zoom.us/j
 /99166955895?pwd=SzI0M3pMVEkvNmw3Q0dqNDVRalZvdz09
LOCATION:Lecture Theatre 2
END:VEVENT
END:VCALENDAR