BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning from humans: A broad overview of approaches to model pref
 erences\, skills and perception - Maria Perez-Ortiz (University of Cambrid
 ge)
DTSTART:20171123T141500Z
DTEND:20171123T151500Z
UID:TALK94744@talks.cam.ac.uk
CONTACT:Alexander Simpson
DESCRIPTION:In different scenarios machine learning algorithms need to lea
 rn from humans. We may want to model preferences to construct a recommende
 r system\, determine the skills of players to better design game tournamen
 ts\, get some insight about our perception of the world or simply because 
 these systems rely on human feedback to interact with a complex environmen
 t. This talk will give a broad introduction to some of the experimental st
 rategies and methods from statistics and machine learning that can be used
  to model human preferences\, skills and perception\, ranging from the use
  of Likert scales\, mean opinion scores and pairwise comparison experiment
 s\, to ranking\, ordinal classification and scaling methods.
LOCATION:SS03 Meeting Room\, Computer Laboratory
END:VEVENT
END:VCALENDAR