Teaching
Bachelor
Master
Master in Brain and Cognitive Sciences - Semester 1, Period 1
Advanced Neural and Cognitive Modelling
This course aims to bring students across Brain and Cognitive Sciences up to speed on some current, exciting computational models in their fields. The course is organized around 5 or 6 recent papers in Brain and Cognitive Science, updated every year. Frequent topics include: deep learning models for predicting brain activity in language processing and vision, dynamical systems models of music processing, reinforcement learning and the cognitive (neuro)science of spatial cognition, non-parametric Bayesian models and artificial language learning, and models of number cognition and the neural basis of mathematics.
Staff: Jelle Zuidema and Ashley Burgoyne
Bachelor Cognition, Language and Communication - Semester 2, Period 1
Advanced Statistics
Building on the course Statistics in R, the goal of this course is to deepen students’ understanding of data analysis methods used in the cognitive and language sciences. After this course, students can apply a number of advanced algorithmic and statistical methods used in classification and regression tasks (supervised learning) and clustering and association task (unsupervised learning).
Master of Logic & Master in Artificial Intelligence - Semester 2, Period 5
Advanced Topics in Computational Semantics
The field of computational semantics is concerned with automatic interpretation of natural language. This course provides an overview of state-of-the-art approaches to language understanding tasks. This is an advanced research seminar aiming to introduce students to recent developments in this field. The course consists of a set of lectures and seminar sessions, where students present and discuss recent research papers. Currently, we focus on representation learning for NLP, considering different levels of language analysis: words, sentences and longer discourse fragments. We also look at the recently proposed contextualised word representation models (such as ELMo and BERT), joint learning methods (including multilingual joint learning and multitask learning) and meta-learning methods (enabling fast model adaptation from only a few examples).
An important component of the course is a research project, in which students have the opportunity to implement a number of semantic models, perform experiments addressing a new research question and write a research paper.
Staff: Ekaterina Shutova
Bachelor Media and Information - Semester 2, Period 1
Coding the Humanities
This course teaches foundational coding skills using Python (a popular programming language) from the perspective of the humanities, with the goals of 1. helping students and researchers to understand when and how to automate a task or analyze data programmatically; and 2. developing custom applications, rather than using ready-made ones, which can benefit the actual practice of humanities research as well as its outputs.
Staff: Jelke Bloem and Rens Bod
Master of Logic & Master in Artificial Intelligence - Semester 1, Period 1
Computational Dialogue Modelling
Conversation or dialogue is the most natural way in which we humans use language, and arguably the holy grail of language-enabled AI systems. With machine learning advancing at such a rapid pace, we now have powerful tools for modelling interacting agents. Yet, to model human-like conversational abilities remains remarkably difficult. This course examines what makes dialogue so challenging, delving into classic and contemporary research in linguistics, cognitive science, NLP and AI.
Staff: Raquel Fernández
Bachelor Media and Information - Semester 2, Period 2
Data-Driven Research
For almost any research question in the humanities, there is a wealth of digital data available. This information goes far beyond the traditional objects of studies in the humanities, and the traditional metadata in heritage or memory institutions (galleries, libraries, archives, museums). How can we make use of the massive amount of content directly, without relying on human description? How to extract meaning from natural language in text? What is the power of semantic representations? How do the emerging AI approaches differ, and can this help promote AI for social good? What research methods help answer these questions? The course has a hands-on approach focusing on assignments and an extensive group research project.
Staff: Jelke Bloem
Master in Artificial Intelligence - Semester 2, Period 5
Deep Learning 2
This course builds upon Machine Learning 1 and Deep Learning 1 to give students access to advanced Deep Learning modelling techniques such as: expressive density estimators (e.g., flow-based models), powerful generative models (e.g., VAE-based and likelihood-free), complex discrete structure, Markov processes and advanced variance reduction techniques, geometric machine learning, temporal machine learning and dynamics, Bayesian NNs and loss-calibrated approximate inference, disentanglement learning, and meta learning.
Staff: Wilker Aziz and Stratis Gavves
Bachelor Media and Information - Semester 2, Period 3
Digital Humanities Lab
In this laboratory course, students engage with a Digital Humanities project. The laboratory allows students to get practical experience on the methods and tools they have explored during previous courses. Students will be able to work on a real-world research project in collaboration with a team or work independently on a topic of choice.
Staff: Giovanni Colavizza
Bachelor Psychobiology - Semester 2, Period 5
Evolution of Language and Music
Most animals produce sounds and communicate, but only humans have music and language. Why do children learn natural language spontaneously and reliably, while other animals fail at even the most basic language tasks? What allows humans to appreciate complex rhythms and melodies, predict from a few notes how a tune will continue, and why has music such profound effects and human emotion? Why do the differences between humans and other animals seem so profound at the level of the music and language system, while we see so many commonalities at the level of perception, pattern recognition and brain functions? In this course we survey different theories of the origins of music and language. We discuss evidence for and against these theories from a variety of sources, including from the archaeological record, linguistic theory, behavioural biology, animal cognition and human and animal genetics. We focus on recent developments in these fields. One such development is a massive effort to gather data on vocal imitation learning in a variety of species, allowing a reconstruction of the evolutionary origins of this ability in birds and mammals (using phylogenetic tree reconstruction algorithms). A second development is a series of computational models and experiments revealing a role for cultural evolution in the evolution of language and music. A third is the discovery of the FoxP2-gene, its supposed role in language and speech, and the variants of this gene in the Neanderthal and songbird genome.
Staff: Jelle Zuidema
Master Brain and Cognitive Sciences - Semester 1, Period 2
Foundations of Neural and Cognitive Modelling
How does the brain, and the networks of spiking neurons it consists of, support the computations required for reasoning, categorisation, vision, language, navigation and many other aspects of human cognition? This course aims at introducing students to the key insights that computational models in the brain and cognitive sciences offer to ultimately answer that question. We discuss multiple modelling paradigms and the relations between them, including: dynamical systems and models of single spiking neurons, McCulloch-Pitts neurons, Hopfields networks, attractor networks, perceptrons, backpropagation, deep learning, reinforcement learning, Bayesian modelling, symbolic models & the binding problem.
Throughout the course, we discuss methodological issues in modelling. The course involves lectures, discussion of literature, pen & paper exercises and a series of computer labs where students, with help of the TA, study key computational models in each of these 6 domains. The courses finishes with student presentations of some key papers, and a miniproject where students do some original modelling.
Staff: Jelle Zuidema
Master in Artificial Intelligence - Semester 2, Period 6
Interpretability & Explainability in AI
The course introduces students to posthoc interpretability techniques, contstrained deep learning and some symbolic/probabilistic Explainability-by-Design approaches. The theory track consists of a series of lecturers, including many guest lectures, and readings. The practice track is organized around a set of workshops that give students much freedom to focus on the subfield of AI they are most interested in.
Staff: Jelle Zuidema and Erman Acar
Bachelor English Language and Culture - Semester 1, Period 1
Introduction to Corpus Linguistics
Corpus linguistics is a methodology whereby large collections of electronically transcribed texts are used in conjunction with computer tools to investigate language. This course aims to provide a general introduction to corpus-based language study. Students will be introduced to different perspectives in the corpus-based analysis of language (qualitative vs. quantitative; diachronic vs. synchronic; monolingual vs. multilingual).
Bachelor Media and Information - Semester 1, Period 1
Introduction to Digital Humanities
The Digital Humanities are increasingly relevant for scholars in all Humanities disciplines. The rapidly rising digital availability of content, as well as the broader digitization of society, offer new exciting opportunities to humanists. This course serves as an introduction to the Cultural Information (IC) track of the Media and Information program, and to the Digital Humanities (DH) as a research area and community.
Staff: Rens Bod
Bachelor Artificial Intelligence - Semester 2, Period 4
Language, Speech and Dialogue Processing
This course provides a comprehensive overview of the fundamentals and most recent approaches to dialogue modeling including natural language and speech processing in dialogue systems. An important part of the course is a collaborative learning environment for students to experiment and apply the theoretical knowledge in practice. The students are expected to practice formulating their own research questions, self-organize, plan and distribute the subtasks to effectively work in a team.
Staff: Wilker Aziz
Master in Artificial Intelligence - Semester 1, Period 2
Natural Language Processing 1
This course introduces the fundamental techniques for a range of tasks in natural language processing (NLP), with a particular focus on statistical and machine learning approaches. We will consider tasks that involve hierarchical structure (e.g., syntactic trees) and/or hidden structure (e.g., in semantic tasks), using supervised and some unsupervised learning algorithms. The course aims to explain the potential and the main limitations of these techniques, as well as discussing them in the wider context of current research issues in NLP and its real-world applications.
An important component of the course is a hands-on practical, in which the students will have the opportunity to implement a number of language processing methods and perform experiments on a real-world task.
Staff: Ekaterina Shutova
Master in Artificial Intelligence - Semester 2, Period 5
Natural Language Processing 2
Do you want to taste the many flavours of cutting-edge research into natural language processing (NLP)? NLP2 is the course that will offer these exciting flavours in the form of mini research projects, pitches by leading researchers and presentations and discussions involving the students and these researchers.
NLP is becoming ever more important and is currently at the heart of AI. This is largely because natural languages are central in daily communication but also because we store our knoweldge, science and history in the form of text and voice recordings. This means that the amount of language data that is available to us electronically is increasing continuously. With this eminent increase, a question arises as to the possibility of inducing latent structure in this data that can be useful discovering how human language understanding works, but also how to use it for engineering tasks such as machine translation, conversational agents and models of human language understanding.
This course covers the breadth of research topics within NLP, considering latent and linguistic structure in language data. Topics include: language modeling, machine translation and paraphrasing, the role of perception (e.g., visual) in language understanding, dialogue systems and interaction, inducing latent, hierarchical or linguistic structure in natural language data, deep generative models, and explanability/interpretability of NLP systems.
Staff: Khalil Sima'an
Bachelor Artificial Intelligence - Semester 2, Period 4
Natuurlijke Taalmodellen en Interfaces
Natural language is the main channel of communication between humans, and much of human knowledge is represented in the form of natural language. Enabling computers to understand it is an extremely important task, and is one of the core problems of artificial intelligence. Though full understanding still remains a remote goal, robust methods have been developed for more shallow forms of processing, and these methods and corresponding formalisms are the focus of this course.
In this course you learn about formalisms and techniques to assign probabilities to (parts of) sentences (language modeling) and to perform basic forms of syntactic and semantic processing. These techniques are the foundation of current data-driven computational linguistics and provide building blocks for speech recognition, language understanding, text summarisation, and machine translation systems.
Staff: Wilker Aziz
Bachelor Artificial Intelligence - Semester 2, Period 5
Taaltheorie & Taalverwerking
Our ability to use natural language to communicate with each other and to record information is one of the main features that makes us intelligent. However, while we use language effortlessly in our everyday life, computers have a hard time processing natural languages such as English or Dutch. Computational linguistics is a subfield of artificial intelligence at the interface of linguistic theory and computer science, which aims at endowing computers with the ability to process natural language. The ultimate goal is to develop artificial agents that can automatically acquire information from text or that can communicate with humans via intelligent interfaces or in human-robot interaction.
This course introduces students to some of the core topics in computational linguistics and natural language processing. We will focus on foundational aspects, paying special attention to rule-based methods. The course provides background for the second-year course Natuurlijke Taalmodellen en Interfaces, which focuses on data-driven probabilistic methods.
The course covers the following key topics in language processing at an introductory level: formal languages and automata, syntactic structure and syntactic parsing, logic-based compositional semantics, word meaning and semantic similarity, vector space models and distributional semantics.
Staff: Sandro Pezzelle
Bachelor Liberal Arts and Sciences, Major Sciences (AUC)
Text Mining
This course provides an introduction to text mining and natural language processing in Python, including real-world applications. The course will introduce fundamental concepts and techniques from computational linguistics, which are used to represent and model texts mathematically. It will also develop on machine learning techniques, including modern deep learning methods, that allow us to make inferences and predictions about, e.g., user experiences, marketing and human behaviour from data that are generated and collected online.
Staff: Jelke Bloem and Giovanni Colavizza