Foundations of Neural and Cognitive Modelling

Course manual 2021/2022

Course content

How do brains implement high-level cognitive functions? How can modelling contribute to answering that question? In this course we consider the conceptual and technical foundations of the major modelling approaches in the brain and cognitive sciences, and explicitly investigate the commonalities and differences. As case studies, we look at models of single neurons (Hodgkin-Huxley, Fitzhugh-Nagumo, McCulloch-Pitts, Rosenblatt), models of networks of neurons (Hopfield, Kohonen, Rumelhart & McClelland, Elman, Hebbian learning, backpropagation), and some basic symbolic and probabilistic models of categorization, reasoning, planning and language (k-means clustering, mixtures of Gaussians, HMM, PCFG). We end with a brief look at how these techniques come together in modern deep learning models.

The lectures combine refreshers on the used mathematical techniques (e.g., ordinary differential equations, vector- and matrix-algebra, probability theory and grammars) with conceptual discussion of the various models, the acceptability of the simplifications they make, and their relations to each other. In the computerlabs we study the properties of the various models (using existing implementations). Towards the end of the course all students present an evaluation of a modelling paper from their own favorite field, discuss its relation to other modelling paradigms and to the modelling methodology discussed in the course.

Objectives

• students are able to explain the conceptual and technical foundations of the major modelling paradigms in brain and cognitive science.
• students can demonstrate their technical skills by solving representative exercises from these fields.
• students can demonstrate their conceptual understanding by critically assessing published models in various subfields, distinguishing between appropriate abstractions and inappropriate simplifications, and by describing relations to models formulated in other paradigms.

Teaching methods

• Lecture
• Computer lab session/practical training
• Seminar
• Presentation/symposium

Lectures, seminars, computer labs

Learning activities

Attendance

Requirements of the programme concerning attendance (OER-B):

1. In the case of practicals, the student must attend at least 80%. Should the student attend less than 80%, he/she must redo the practical, or the Examinations Board may have one or more supplementary assignments issued.
2. In the case of study-group sessions with assignments, the student must attend at least 80% of the study-group sessions. Should the student attend less than 80%, he/she must redo the study group, or the Examinations Board may have one or more supplementary assignments issued.

Assessment

Assignments

Assignment 1,2,3,4,5,6

• Weekly assignments

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Course structure

Timetable

The schedule for this course is published on DataNose.

Additional information

More information on the 2019-2020 edition of the course, including more details on the recommended prior knowledge, can be found via: http://projects.illc.uva.nl/LaCo/computationallinguistics/

Students taking this course need to have a strong interest in modelling and brain and cognitive science, and some talent for mathematical thinking. The course is also open for motivated students in artificial intelligence and logic. Contact the lecturer (zuidema@uva.nl) when in doubt.

Last year's course evaluation

In order to provide students some insight how we use the feedback of student evaluations to enhance the quality of education, we decided to include the table below in all course guides.

Contact information

Coordinator

• dr. Jelle Zuidema