Dieuwke Hupkes bio photo

Dieuwke Hupkes

Research scientist at Meta, ELLIS Scholar

Email Twitter Instagram Github

Curriculum Vitae


Teaching

Overview

I obtained the Dutch Basic Teaching Qualification (UTQ/BKO) at the University of Amsterdam.

Courses

March - July 2020
Taaltheorie en Taalverwerking (Universiteit van Amsterdam)
First year Bsc AI course that introduces key topic in natural language processing at an introductory level.
March - July 2020
Human(e) AI (Universiteit van Amsterdam)
Elective Bsc coure about the role of AI in society.

Thesis supervision


April - June 2020
Hugh Mee Wong, Bachelor Artificial Intelligence
Subject-Verb Number Agreement in a Dutch LSTM Language Model
April - June 2020
Jeroen Taal, Bachelor Artificial Intelligence
Comparing encoding of number information in dense and sparse LSTMs
April - August 2020
Oskar van der Wal, Master Artificial Intelligence
The Grammar of Emergent Languages
Co-supervision with Elia Bruni
April - Now
Lewis O'Sullivan, Master Brain and Cognitive Science
Co-supervision with Willem Zuidema
June 2019 - May 2020
Oscar Ligthart, Master Artificial Intelligence
Measuring Compositionality in Agent Emerged Languages
Co-supervision with Elia Bruni
June 2018 - Feb 2020



Jan - Aug 2019



Jan - Aug 2019



Jan - July 2019



Jan - Nov 2018



Jan - Oct 2018




Jan - Aug 2018



Jan - Aug 2018



Jan - July 2018



Jan - Aug 2018



Jan - Jun 2018



Spring 2017

Jaap Jumelet, Master Artificial Intelligence
Interpreting Language Models
Co-supervision with Willem Zuidema

Gautier Dagan, Master Artificial Intelligence
Co-evolution of language and agent in referential games
Co-supervision with Elia Bruni

Dennis Ulmer, Master Artificial Intelligence
Recoding latent sentence representations
Co-supervision with Elia Bruni

Diana Rodriguez Luna, Master Artificial Intelligence
Language emregence in multi-agent referential games
Co-supervision with Elia Bruni

Kris Korrel, Master Artificial Intelligence
From sequence to attention
Co-supervision with Elia Bruni and Germán Kruszewski

Sanne Bouwmeester, Master Artificial Intelligence
Analysing seq-to-seq models in goal-oriented dialogue: generalising to disfluencies
Co-supervision with Raquel Fernandez

Krsto Proroković, Master of Logic
Learning to decide a formal language: a recurrent neural network approach
Co-supervision with Elia Bruni and Germán Kruszewski

Anand Kumar Singh, Master Computational Intelligence
Pondering in artificial neural networks
Co-supervision with Elia Bruni and Germán Kruszewski

Ujjwal Sharma, Master Artificial Intelligence
Interpreting decision-making in interactive visual dialogue
Co-supervision with Elia Bruni

Rezka Aufar Leonandya, Master Artificial Intelligence
Learning to follow instructions
Co-supervision with Elia Bruni and Germán Kruszewski

Lucas Weber, Master Brain and Cognitive Science
Continual learning in humans and neuroscience-inspired AI
Co-supervision with Elia Bruni

Philip Bouman, Bachelor Artificial Intelligence Thesis
Modelling fonts with convolutional neural networks.
Co-supervision with Jelle Zuidema

Individual/group projects

Feb 2020 - now


Feb 2020 - now


June 2019



June 2019


June 2018



June 2018


June 2018


January 2018

Contextual decomposition for attention-based models
Msc AI student Tom Kersten

Curriculum learning to improve compositionality
Msc AI student Michael Neely

Syntactic awareness in language models: recurrence vs self-attention
Msc AI students Sander Bos, Lorian Colthof, Bryan Guevara and Vivian van Oijen

Unsupervised grammar induction in emergent languages
Msc AI students Silvan de Boer and Oskar van der Wal

On the realisation of compositionality in neural networks
Msc AI students Joris Baan, Jana Leible, Mitja Nikolaus, David Rau, Verna Dankers, Snthosh Rajamanickam and Dennis Ulmer

Analysing subject-verb agreement with diagnostic classification
Msc AI and Logic students Mario Giulianelli, Jack Harding and Florian Mohnert

What do language models encode?
Msc AI student Jaap Jumelet

Learning compositionality in neural networks
Master of Logic students Federico Schiaffino, Haukur Pál Jonnson, Max Rapp, Flavio Tisi and Yuan-Ho Yao

(Guest) lectures

I’ve (guest)lectured in the following courses:

April 2020
May 2019
Nov 2018
Dec 2017
Mar 2017

Natural Language Processing 2
Statistical Methods for Natural Language Semantics
Foundations of Neural and Cognitive Modelling
Natural Language Processing 1
Cognitive Models of Language and Music

Teaching assistant


I have been a teaching assistant in the following courses:

Oct-Dec 2017




Sept-Oct 2016




Feb-April 2016
Oct-Dec 2016



Oct-Dec 2015




Feb-Apr 2014
Feb-Apr 2015


Apr-Jun 2013
April-Jun 2012




Feb 2014
Feb 2013
Feb 2012


Feb-Jun 2013

Natural Language Processing 1 (Universiteit van Amsterdam)
This course aims at providing the student with the background that is needed for studying statistical models that are used in the field of Computational Linguistic, taught in the Master AI and the Master of Logic

Computational Semantics and Pragmatics (Universiteit van Amsterdam)
Course on topics and methodologies of natural language semantics and pragmatics from an empirical point of view, taught in the Master AI and Master of Logic.

Evolution of Language and Music (Universiteit van Amsterdam)
Bachelor's course on the main theories about the origins and evolution of language and music, including computational analysis on data relevant to this topic.

Foundations of Neural and Cognitive Modelling (Universiteit van Amsterdam)
Master course on conceptual and technical foundations of the major modelling paradigms in brain and cognitive science.

Unsupervised Language Learning (Universiteit van Amsterdam)
Obligatory course on techniques in Natural Language Processing for first year Master’s students in Artificial Intelligence in the language Track.

Automata and Formal Languages (Universiteit van Amsterdam)
Course for first year undergraduate students in computer science. Introduction to the theory of formal languages and automata. Among included topics are finite state automata, push-down automata, regular and context free languages, Turing machines and computability.

Logic (Universiteit van Amsterdam)
Introduction to logic for first years students of the bachelor Beta-gamma. Topics taught include standard classical logic, logical transcription of information and elementary methods for testing validity.

Biomechanics (Vrije Universiteit)
Course on the principles of classical Newtonian mechanics applied to the human body for first years Bewegingswetenschappen students.