BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Adaptive Group Sparsity Using the Ordered Weighted l1 Regularizer 
 - Prof. Mario Figueiredo\, Instituto de Telecomunicações and Instituto S
 uperior Técnico University of Lisbon\, Portugal
DTSTART:20160526T140000Z
DTEND:20160526T150000Z
UID:TALK66300@talks.cam.ac.uk
CONTACT:Prof. Ramji Venkataramanan
DESCRIPTION:In high-dimensional linear regression (and other inverse probl
 ems)\,\nhighly correlated variables/covariates are a challenge to variable
  selection\nprocedures. In those scenarios\, standard sparsity-inducing re
 gularization\n(namely l1\, a.k.a. LASSO) is inadequate\, as it leads to th
 e selection of\narbitrary convex combinations of those variables\, maybe e
 ven of an\narbitrary subset thereof. However\, specially in scientific con
 texts\, it is\noften important to explicitly identify all the relevant cov
 ariates\, as well as\nexplicitly identify groups/clusters thereof. This ta
 lk addresses the recently\nintroduced ordered weighted l1 regularizer\, wh
 ich has been proposed for\nthis purpose. We review several convex optimiza
 tion aspects concerning\nthis regularizer\, show sufficient conditions for
  exact clustering\,\nand characterize its statistical performance.\n\n\n*B
 io*: Mário A. T. Figueiredo received a PhD in electrical and\ncomputer en
 gineering\, from Instituto Superior Técnico (IST)\, the\nengineering scho
 ol of the University of Lisbon\, in 1994. He has been\nwith the faculty of
  the Department of Electrical and Computer\nEngineering\, IST\, since 1994
 \, where he is now a Professor.\nHe is also area coordinator and group lea
 der at Instituto de\nTelecomunicações\, a private non-profit research in
 stitute. His\nresearch interests include image processing and analysis\, m
 achine\nlearning\, and optimization. M. Figueiredo is a Fellow of the IEEE
 \nand of the IAPR and is included in the 2014 and 2015 Thomson Reuters'\nH
 ighly Cited Researchers lists\; he received several awards\, namely\nthe 2
 011 IEEE Signal Processing Society Best Paper Award\, the\n2014 IEEE W. R.
  G. Baker Award\, and the 2016 EURASIP Individual\nTechnical Achievement A
 ward.\n
LOCATION:LR6\, Department of Engineering
END:VEVENT
END:VCALENDAR