BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Modelling the Human Eye: A look at statistical and machine learnin
 g techniques - Adam Kashlak (CCA)
DTSTART:20160309T160000Z
DTEND:20160309T170000Z
UID:TALK65046@talks.cam.ac.uk
CONTACT:Mr Simone Parisotto
DESCRIPTION:Modelling and predicting the complex movements of the human ey
 e when it is presented with an image to observe has been a challenging tas
 k for vision researchers\, computer scientists\, and statisticians.  In 20
 15\, the Royal Statistical Society presented a data set of eye movements a
 s a challenge problem to young researchers.  Using this data as a foundati
 on\, we will present an expository look at a variety of machine learning a
 nd statistics techniques in the context of this data.  Methods include spa
 tial clustering algorithms\, nearest-neighbour classifiers\, non-parametri
 c density estimation\, mixture models\, Bayesian model selection\, and (no
 t-quite-hidden) Markov models.  There will also be a lot of pretty picture
 s.
LOCATION:MR14\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR