BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Resource Allocation for Statistical Estimation - Quentin Berthet\,
  California Institute of Technology
DTSTART:20150309T160000Z
DTEND:20150309T170000Z
UID:TALK58257@talks.cam.ac.uk
CONTACT:20082
DESCRIPTION:Statistical estimation in many contemporary settings involves 
 the acquisition\, analysis\, and aggregation of datasets from multiple sou
 rces\, which can have significant differences in character and in value. D
 ue to these variations\, the effectiveness of employing a given resource 
 – e.g.\, a sensing device or computing power – for gathering or proces
 sing data from a particular source depends on the nature of that source. A
 s a result\, the appropriate division and assignment of a collection of re
 sources to a set of data sources can substantially impact the overall perf
 ormance of an inferential strategy. We adopt a general view of the notion 
 of a resource and its effect on the quality of a data source\, and we desc
 ribe a framework for the allocation of a given set of resources to a colle
 ction of sources in order to optimize a specified metric of statistical ef
 ficiency. We discuss several stylized examples involving inferential tasks
  such as parameter estimation and hypothesis testing based on heterogeneou
 s data sources\, in which optimal allocations can be computed either in cl
 osed form or via efficient numerical procedures based on convex optimizati
 on.
LOCATION:MR12\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR