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:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20191103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:69f37787ba37a
DTSTART;TZID=America/Toronto:20200110T100000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20200110T100000
URL:https://uwaterloo.ca/statistics-and-actuarial-science/events/department
 -seminar-fangzheng-xie-johns-hopkins-university
LOCATION:M3 - Mathematics 3 200 University Avenue West Room 3127 Waterloo O
 N N2L 3G1 Canada
SUMMARY:Department seminar by Fangzheng Xie\, Johns Hopkins University
CLASS:PUBLIC
DESCRIPTION:GLOBAL AND LOCAL ESTIMATION OF LOW-RANK RANDOM GRAPHS\n\n------
 -------------------\n\nRandom graph models have been a heated topic in sta
 tistics and machine\nlearning\, as well as a broad range of application ar
 eas. In this talk\nI will give two perspectives on the estimation task of 
 low-rank random\ngraphs. Specifically\, I will focus on estimating the lat
 ent positions\nin random dot product graphs. The first component of the ta
 lk focuses\non the global estimation task. The minimax lower bound for glo
 bal\nestimation of the latent positions is established\, and this minimax\
 nlower bound is achieved by a Bayes procedure\, referred to as the\nposter
 ior spectral embedding. The second component of the talk\naddresses the lo
 cal estimation task. We define local efficiency in\nestimating each indivi
 dual latent position\, propose a novel one-step\nestimator that takes adva
 ntage of the curvature information of the\nlikelihood function (i.e.\, de
 rivatives information) of the graph\nmodel\, and show that this estimator
  is locally efficient. The\npreviously widely adopted adjacency spectral e
 mbedding is proven to be\nlocally inefficient due to the ignorance of the 
 curvature information\nof the likelihood function. Simulation examples and
  the analysis of a\nreal-world Wikipedia graph dataset are provided to dem
 onstrate the\nusefulness of the proposed methods.
DTSTAMP:20260430T153847Z
END:VEVENT
END:VCALENDAR