BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:TSE
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
RDATE:20160327T020000
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.81575.field_date.0@www.tse-fr.eu
DTSTAMP:20260616T075751Z
CREATED:20150724T081001Z
DESCRIPTION:Alexandre Tsybakov (ENSAE)\, “Oracle inequalities for network m
 odels and sparse graphon estimation”\, Statistics Seminar\, Toulouse: TSE\
 , October 1\, 2015\, 11:00–12:30\, room MS 001.\n\nInhomogeneous random gr
 aph models encompass many network models such as stochastic block models a
 nd latent position models. In this paper\, we study two estimators – the o
 rdinary block constant least squares estimator\, and its restricted versio
 n. We show that they satisfy oracle inequalities with respect to the block
  constant oracle. As a consequence\, we derive optimal rates of estimation
  of the probability matrix. Our results cover the important setting of spa
 rse networks. Nonparametric rates for graphon estimation in the L2 norm ar
 e also derived when the probability matrix is sampled according to a graph
 on model. The results shed light on the differences between estimation und
 er the empirical loss (the probability matrix estimation) and under the in
 tegrated loss (the graphon estimation).
DTSTART;TZID=Europe/Paris:20151001T110000
DTEND;TZID=Europe/Paris:20151001T123000
LAST-MODIFIED:20170723T051233Z
LOCATION:Toulouse: TSE\, October 1\, 2015\, 11:00–12:30\, room MS 001
SUMMARY:Statistics Seminar
URL;TYPE=URI:https://www.tse-fr.eu/seminars/2015-oracle-inequalities-networ
 k-models-and-sparse-graphon-estimation
END:VEVENT
END:VCALENDAR