BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/522
DTSTAMP:20230914T125927Z
SUMMARY:Testing the Manifold Hypothesis
DESCRIPTION:Speaker: Hariharan Narayanan (Universithy of Washington\nDepart
ment of Statistics and\nDepartment of Mathematics\nPadelford Hall\, Room C
-301\nSeattle\, WA 98105\nUnited States of America)\n\nAbstract: \nAbstrac
t: We are confronted with very high dimensional data sets. As a result\, m
ethods of dealing with high dimensional data have become prominent. One ge
ometrically motivated approach for analyzing data is called manifold learn
ing. The underlying hypothesis of this subfield of machine learning is tha
t high dimensional data tend to lie near a low dimensional manifold. Howev
er\, the basic question of understanding when data lies near a manifold is
poorly understood. I will describe joint work with Charles Fefferman and
Sanjoy Mitter on developing a provably correct algorithm to test this hypo
thesis using i.i.d samples from an arbitrary distribution supported in the
unit ball in a Hilbert space.\n
URL:https://www.tcs.tifr.res.in/web/events/522
DTSTART;TZID=Asia/Kolkata:20140804T143000
DTEND;TZID=Asia/Kolkata:20140804T153000
LOCATION:AG-69
END:VEVENT
END:VCALENDAR