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:Machine Learning Reading Group @ CUED
SUMMARY:Kernel Mean Embeddings - Elre Oldewage (University
of Cambridge)
DTSTART;TZID=Europe/London:20200219T110000
DTEND;TZID=Europe/London:20200219T123000
UID:TALK139993AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/139993
DESCRIPTION:Kernel mean embeddings are a technique that repres
ents probability distributions as elements in a hi
gh dimensional Hilbert space. This allows difficul
t operations such as expectations to be reformulat
ed as dot products in the Hilbert space.\nThis tal
k introduces reproducing kernel Hilbert spaces (RK
HS)\, which is the theoretical underpinnings upon
which kernel mean embeddings are constructed. The
talk aims to provide intuition regarding the relat
ionship between kernels\, feature maps and RHKS fu
nction spaces. The talk continues with a discussio
n of kernel mean embeddings\, and common applicati
ons of kernel mean embeddings\, specifically\, ker
nel two-sample hypothesis testing. We also discuss
embedding multivariable conditional distributions
which allow the application of kernel Bayes rule.
LOCATION:Engineering Department\, CBL Room BE-438
CONTACT:
END:VEVENT
END:VCALENDAR