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:Inference Group
SUMMARY:Monte Carlo is Bayesian - John Skilling\, ex-DAMTP
DTSTART;TZID=Europe/London:20051109T150000
DTEND;TZID=Europe/London:20051109T160000
UID:TALK4498AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/4498
DESCRIPTION:Cox was right\, Kolmogorov was wrong. The attempt
to found probability calculus upon set theory see
ms to be misguided\, because it is necessary to ex
clude sets of measure zero in order to avoid parad
ox. It is better to start with unit measure of pr
obabilistic belief\, to be distributed among relev
ant hypotheses regardless of any measure they may
possess. This improved viewpoint shows that\, con
trary to the folklore of the subject\, Monte Carlo
integration is properly probabilistic (as an algo
rithm for Bayesian computation should be). Nested
sampling is an extension to problems of larger sc
ale: it is an algorithm of wider scope that can de
al with a variety of multi-modal problems better t
han conventional annealing.
LOCATION:Ryle Seminar Room\, Cavendish Laboratory
CONTACT:Phil Cowans
END:VEVENT
END:VCALENDAR