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/2015/newsitem/7017/30- October-2015-Master-of-AI-defense-Sara-Veldhoen DTSTAMP:20151011T000000 SUMMARY:Master of AI defense, Sara Veldhoen DTSTART;TZID=Europe/Amsterdam:20151030T140000 DTEND;TZID=Europe/Amsterdam:20151030T000000 LOCATION:Room F1.15, ILLC, Science Park 107, Amste rdam ATTENDEE;ROLE=Supervisor:Jelle Zuidema DESCRIPTION:I have investigated the appropriatenes s for natural languge inference of the distributed representations that result from a neural network -based approach to compositional distributed seman tics. I found that we have to be very careful in d rawing conclusions, and present some recommendatio ns to better assess the adequacy of the models for this task. For more information, please contact sara.veldhoen at student.uva.nl X-ALT-DESC;FMTTYPE=text/html:\n
I have i nvestigated the appropriateness for natural langug e inference of the distributed representations tha t result from a neural network-based approach to c ompositional distributed semantics. I found that w e have to be very careful in drawing conclusions, and present some recommendations to better assess the adequacy of the models for this task.
\n \n For more information, please contact sara.veldhoen at span> student.uva.nl URL:/NewsandEvents/Archives/2015/newsitem/7017/30- October-2015-Master-of-AI-defense-Sara-Veldhoen END:VEVENT END:VCALENDAR