BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:ML@CL Seminar Series
SUMMARY:Bayesian Inference for Optimal Transport with Stoc
hastic Cost - Anton Mallasto
DTSTART;TZID=Europe/London:20210330T150000
DTEND;TZID=Europe/London:20210330T160000
UID:TALK158512AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/158512
DESCRIPTION:In machine learning and computer vision\, optimal
transport (OT) has had significant success in lear
ning generative models and defining metrics betwee
n structured and stochastic data objects\, that ca
n be cast as probability measures. The key element
of optimal transport is the so called lifting of
an \\emph{exact} cost (distance) function\, define
d on the sample space\, to a cost (distance) betwe
en probability measures over the sample space. Thi
s is carried out by minimizing the total transport
ation cost between two measures\, resulting in the
OT plan.\n\nHowever\, in many real life applicati
ons the cost is stochastic: for example\, an unpre
dictable traffic flow affects the cost of transpor
tation between a factory and an outlet. In this ta
lk\, we devise a Bayesian approach for inferring a
distribution over the OT plans in such random set
tings.
LOCATION:https://us02web.zoom.us/j/88368107456?pwd=dTkvaXBa
aCszMnB4ck1CRXNXVWtTQT09
CONTACT:
END:VEVENT
END:VCALENDAR