BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260610T081243EDT-8911UBuZrD@132.216.98.100
DTSTAMP:20260610T121243Z
DESCRIPTION:Distributed Kernel Regression for Large-scale Data\n\n\n	\n		\n			\n				
 \n					In modern scientific research\, massive datasets with huge numbers of ob
 servations are frequently encountered. To facilitate the computational pro
 cess\, a divide-and-conquer scheme is often used for the analysis of big d
 ata. In such a strategy\, a full dataset is first split into several manag
 eable segments\; the final output is then aggregated from the individual o
 utputs of the segments. Despite its popularity in practice\, it remains la
 rgely unknown that whether such a distributive strategy provides valid the
 oretical inferences to the original data\; if so\, how efficient does it w
 ork? In this talk\, I address these fundamental issues for the non-paramet
 ric distributed kernel regression\, where accurate prediction is the main 
 learning task. I will begin with the naive simple averaging algorithm and 
 then talk about an improved approach via ADMM. The promising preference of
  these methods is supported by both simulation and real data examples.\n			\n
 		\n	\n\n
DTSTART:20170314T150000Z
DTEND:20170314T150000Z
LOCATION:room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Chen Xu
URL:https://www.mcgill.ca/mathstat/channels/event/chen-xu-267013
END:VEVENT
END:VCALENDAR