BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:8202@i2m.univ-amu.fr
DTSTART;TZID=Europe/Paris:20141219T140000
DTEND;TZID=Europe/Paris:20141219T150000
DTSTAMP:20241120T210105Z
URL:https://www.i2m.univ-amu.fr/evenements/raphael-bailly-utc-tensor-facto
 rization-for-multi-relational-learning/
SUMMARY: (...): Raphael Bailly (UTC): Tensor factorization for multi-relati
 onal learning
DESCRIPTION:: Title: Tensor factorization for multi-relational learning\n\n
 Abstract:\nLearning relational data has been of a growing interest in fiel
 ds as diverse as modeling social networks\\\, semantic web\\\, or bioinfor
 matics. To some extent\\\, a network can be seen as multi-relational data\
 \\, where a particular relation represents a particular type of link betwe
 en entities. It can be modeled as a three-way tensor.\n\nTensor factorizat
 ion have shown to be a very efficient way to learn such data. It can be do
 ne either in a 3-way factorization style (trigram\\\, e.g. RESCAL) or by s
 um of 2-way factorization (bigram\\\, e.g TransE). Those methods usually a
 chieve state-of-the-art accuracy on benchmarks. Though\\\, all those learn
 ing methods suffer from regularization processes which are not always adeq
 uate.\n\nWe show that both 2-way and 3-way factorization of a relational t
 ensor can be formulated as a simple matrix factorization problem. This cla
 ss of problems can naturally be relaxed in a convex way. We show that this
  new method outperforms RESCAL on two benchmarks.\n
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20141026T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR