BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.2//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:Eventi DIAG
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:STANDARD
DTSTART:20161030T030000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20160327T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RDATE:20170326T020000
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.7374.field_data.0@www.ugov-ricerca.uniroma1.it
DTSTAMP:20260411T042105Z
CREATED:20160929T090915Z
DESCRIPTION:In this talk I will introduce the Inter-Battery Topic Model (IB
 TM). Our approach extends traditional topic models by learning a factorize
 d latent variable representation. The structured representation leads to a
  model that marries benefits traditionally associated with a discriminativ
 e approach\, such as feature selection\, with those of a generative model\
 , such as principled regularization and ability to handle missing data. Th
 e factorization is provided by representing data in terms of aligned pairs
  of observations as different views. This  provides means for selecting a 
 representation that separately models topics that exist in both views from
  the topics that are unique to a single view. This structured consolidatio
 n allows for efficient and robust inference and provides a compact and eff
 icient representation.BioHedvig Kjellström is a Professor of Computer Scie
 nce and the head of the Computer Vision and Active Perception Lab (CVAP) a
 t KTH in Stockholm\, Sweden. She received an MSc in Engineering Physics an
 d a PhD in Computer Science from KTH in 1997 and 2001\, respectively. The 
 topic of her doctoral thesis was 3D reconstruction of human motion in vide
 o. Between 2002 and 2006 she worked as a scientist at the Swedish Defence 
 Research Agency\, where she focused on Information Fusion and Sensor Fusio
 n. In 2007 she returned to KTH\, pursuing research in activity analysis in
  video. Her present research focuses on the modeling of perception and pro
 duction of human non-verbal communicative behavior and activity\, with app
 lications in Health\, Robotics\, and Performing Arts.In 2010\, she was awa
 rded the Koenderink Prize for fundamental contributions in Computer Vision
  for her ECCV 2000 article on human motion reconstruction\, written togeth
 er with Michael Black and David Fleet. She has written around 70 papers in
  the fields of Robotics\, Computer Vision\, Information Fusion\, Machine L
 earning\, Cognitive Science\, Speech\, and Human-Computer Interaction. She
  is mostly active within the areas of Robotics and Computer Vision\, where
  she is an Associate Editor for IEEE TPAMI and IEEE RA-L\, and an Area Cha
 ir for CVPR 2016 and RSS 2016.
DTSTART;TZID=Europe/Paris:20160929T110000
DTEND;TZID=Europe/Paris:20160929T110000
LAST-MODIFIED:20190805T155749Z
LOCATION:Room B 101
SUMMARY:Prof. H. Kjellstrom: Learning Factorized Latent Representations Usi
 ng IBTM - Prof. H. Kjellstrom\, KTH Stockholm
URL;TYPE=URI:http://www.ugov-ricerca.uniroma1.it/node/7374
END:VEVENT
END:VCALENDAR