BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:On the Computational and Statistical Interface and &quot\;Big Data
 &quot\; - Michael I. Jordan\, University of California\, Berkeley
DTSTART:20130510T144500Z
DTEND:20130510T154500Z
UID:TALK45237@talks.cam.ac.uk
CONTACT:Richard Samworth
DESCRIPTION:The rapid growth in the size and scope of datasets in science 
 and technology\nhas\ncreated a need for novel foundational perspectives on
  data analysis that\nblend\nthe statistical and computational sciences.  T
 hat classical perspectives\nfrom these fields are not adequate to address 
 emerging problems in "Big\nData" is apparent\nfrom their sharply divergent
  nature at an elementary level---in computer\nscience\, the growth of the 
 number of data points is a source of "complexity"\nthat must be tamed via 
 algorithms or hardware\, whereas in statistics\, the\ngrowth\nof the numbe
 r of data points is a source of "simplicity" in that inferences\nare gener
 ally stronger and asymptotic results can be invoked.  Indeed\, if\ndata ar
 e a statistician's principal resource\, why should more data be\nburdensom
 e\nin some sense?  Shouldn't it be possible to exploit the increasing\ninf
 erential strength of data at scale to keep computational complexity at\nba
 y?  I present three research vignettes that pursue this theme\, the first\
 ninvolving the deployment of resampling methods such as the bootstrap on\n
 parallel and distributed computing platforms\, the second involving\nlarge
 -scale matrix completion\, and the third introducing a methodology of\n"al
 gorithmic weakening\," whereby hierarchies of convex relaxations are used\
 nto control statistical risk as data accrue.\n\nJoint work with Venkat\nCh
 andrasekaran\, Ariel Kleiner\, Lester Mackey\, Purna Sarkar\, and Ameet\nT
 alwalkar
LOCATION:Seminar Room 1\, Isaac Newton Institute
END:VEVENT
END:VCALENDAR