BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Statistical Relational Learning: Review and Recent Advances - Lise
  Getoor (University of California\, Santa Cruz)
DTSTART:20160725T143000Z
DTEND:20160725T150000Z
UID:TALK66841@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Statistical relational learning (SRL) is a subfield of machine
  learning that combines relational representations (from databases and AI)
  with probabilistic modeling techniques (most often graphical models)for m
 odeling network data (typically richly structured multi-relational and mul
 ti-model networks). &nbsp\;In this talk\, I will briefly review some SRL m
 odeling techniques\, and then I will introduce hinge-loss Markov random fi
 elds (HL-MRFs)\, a new kind of probabilistic graphical model that supports
  scalable collective inference from richly structured data. &nbsp\;HL-MRFs
  unify three different approaches to convex inference: LP approximations f
 or randomized algorithms\, local relaxations for probabilistic graphical m
 odels\, and inference in soft logic. &nbsp\;I will show that all three lea
 d to the same inference objective. &nbsp\;This makes inference in HL-MRFs 
 highly scalable. &nbsp\; Along the way\, I will describe several successfu
 l applications of HL-MRFs and I will describe probabilistic soft logic\, a
  declarative language for defining HL-MRFS.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR