BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20151025T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20150329T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.7082.field_data.0@www.glad.uniroma1.it
DTSTAMP:20260405T093559Z
CREATED:20150616T133136Z
DESCRIPTION:In this talk\, we describe a general algorithmic approach to no
 nparametric learning in data streams.  Our method covers the input space u
 sing simple classifiers that are locally trained. A good balance between m
 odel complexity and predictive accuracy is achieved by dynamically adaptin
 g the cover to the local complexity of the classification problem. For the
  simplest instance of our approach\, we prove a theoretical performance gu
 arantee against any Lipschitz classifier and without stochastic assumption
 s on the stream. Experiments on standard benchmarks complement the theoret
 ical results\, showing good performance even when the model size is kept i
 ndependent of the stream length. 
DTSTART;TZID=Europe/Paris:20150622T113000
DTEND;TZID=Europe/Paris:20150622T113000
LAST-MODIFIED:20150621T123344Z
LOCATION:Room B203
SUMMARY:An algorithmic approach to nonparametric online learning - Prof Nic
 olo' Cesa-Bianchi\, University of Milan La Statale
URL;TYPE=URI:http://www.glad.uniroma1.it/node/7082
END:VEVENT
END:VCALENDAR