BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20170312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20161106T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:69b071d4d482d
DTSTART;TZID=America/Toronto:20170726T100000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20170726T100000
URL:https://uwaterloo.ca/artificial-intelligence-group/events/masters-thesi
 s-presentation-naive-bayes-data-complexity-and
LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven
 ue West 2306C Waterloo ON N2L 3G1 Canada
SUMMARY:Master's Thesis Presentation: Naive Bayes Data Complexity and\nChar
 acterization of Optima of the Unsupervised Expected Likelihood
CLASS:PUBLIC
DESCRIPTION:Speaker: Ali Wytsma\, Master's Candidate\n\nThe naive Bayes mod
 el is a simple model that has been used for many\ndecades\, often as a bas
 eline\, for both supervised and unsupervised\nlearning. With a latent clas
 s variable it is one of the simplest\nlatent variable models\, and is ofte
 n used for clustering. The\nestimation of its parameters by maximum likeli
 hood (e.g.\, gradient\nascent\, expectation maximization) is subject to lo
 cal optima since the\nobjective is non-concave. However\, the conditions u
 nder which global\noptimality can be guaranteed are currently unknown.
DTSTAMP:20260310T193236Z
END:VEVENT
END:VCALENDAR