BEGIN:VCALENDAR
VERSION:2.0
PRODID:ILLC Website
X-WR-TIMEZONE:Europe/Amsterdam
BEGIN:VTIMEZONE
TZID:Europe/Amsterdam
X-LIC-LOCATION:Europe/Amsterdam
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:/NewsandEvents/Archives/2017/newsitem/8515/19-
 January-2017-Computational-Linguistics-Seminar-Gem
 ma-Boleda
DTSTAMP:20170109T141411
SUMMARY:Computational Linguistics Seminar, Gemma B
 oleda
ATTENDEE;ROLE=Speaker:Gemma Boleda (Pompeu Fabra, 
 Barcelona)
DTSTART;TZID=Europe/Amsterdam:20170119T160000
LOCATION:Room F1.15, Science Park 107, Amsterdam
DESCRIPTION:Abstract:  Over a century ago, Frege f
 amously introduced the distinction between sense a
 nd reference that is one of the theoretical founda
 tions of formal semantics. However, in practice fo
 rmal semanticists took reference and ran away with
  it, either eschewing sense-related issues altoget
 her or giving a referential treatment to them (wit
 h notable exceptions). In this talk, I argue that 
 we need to go back to Fregean sense, and propose t
 hat data-induced, continuous representations provi
 ded by distributional semantics and deep learning 
 methods provide a good methodological handle for s
 ense-related aspects of meaning. I support these c
 laims with results from both computational modelin
 g and theoretical studies. I then revisit referenc
 e and present ongoing work on the challenging ente
 rprise of tackling it with continuous methods, too
 .
X-ALT-DESC;FMTTYPE=text/html:\n  <p>Abstract:<br>\
 n  Over a century ago, Frege famously introduced t
 he distinction between sense and reference that is
  one of the theoretical foundations of formal sema
 ntics. However, in practice formal semanticists to
 ok reference and ran away with it, either eschewin
 g sense-related issues altogether or giving a refe
 rential treatment to them (with notable exceptions
 ). In this talk, I argue that we need to go back t
 o Fregean sense, and propose that data-induced, co
 ntinuous representations provided by distributiona
 l semantics and deep learning methods provide a go
 od methodological handle for sense-related aspects
  of meaning. I support these claims with results f
 rom both computational modeling and theoretical st
 udies. I then revisit reference and present ongoin
 g work on the challenging enterprise of tackling i
 t with continuous methods, too.</p>\n
URL:https://www.illc.uva.nl/LaCo/CLS/
END:VEVENT
END:VCALENDAR