utl-test

Events

3 - 5 June 2019

UTL Lectures: Aurelie Herbelot

We are very pleased to welcome Aurelie Herbelot  from the University of Trento who is visiting UiL-OTS to give a series of 2 lectures plus a seminar, from June 3-5.

 

Crash course in Computational Semantics, June 3-5

Aurelie Herbelot , Univ. of Trento

 

Lecture 1 – Distributional semantics as lexical representation

June 3 (Mon) : 11:00-12:45 : JK 2-3 – 109 

This lecture will give an introduction to distributional semantics as a representation of the lexicon. Core techniques will be introduced to understand ‘traditional’ statistical models as well as their neural network counterparts. Evaluation strategies will be presented, both for the word and phrase level.

* Lecture 2 – Distributional semantics beyond the lexicon

June 4 (Tue)  :  11:00-12:45: D25 – 003

This lecture will introduce techniques for modeling aspects of language beyond the lexicon. We will cover three topics: a) compositionality, both in traditional and neural models; b) the representation of function words; c) the relation of distributional models to the world, both at a conceptual and perceptual level.

* Seminar – Speaker-dependence in distributional semantics

June 5 (Wed) :  11:00-12:45 : JK 2-3 – 019

One long-standing puzzle in semantics is the ability of speakers to refer successfully in spite of holding different models of the world. This puzzle is famously illustrated by the cup/mug example: if two speakers disagree on whether a specific entity is a cup or a mug (i.e. if their interpretation functions differ), how can they align so that the entity can still be talked about? Another puzzle, coming to us through lexical and distributional semantics, is that word meaning seems to be infinitely flexible across utterances, indeed much more so than the traditional notion of sense would have it. This makes the alignment process between speakers even more unpredictable. In this talk, I will report on a series of experiments aiming at investigating differences in language use through distributional semantics techniques. I will sketch what such differences can tell us about the ability of speakers to align at a model-theoretic level.

 

Attendance is free and all students and colleagues are very welcome, but please register if you can at https://forms.gle/cHm3BUr87wFEx7177

 

You can find more about Aurelie Herbelot’s research on her webpage: https://aurelieherbelot.net