Machine learning for cognitive computational neuroscience

DozentIn: Prof. Dr. rer. nat. Tim Christian Kietzmann , M. Sc. Philip Sulewski

Veranstaltungstyp: Vorlesung mit Seminar

Ort: 66/E34: Di. 10:00 - 12:00 (14x), 93/E33: Do. 12:00 - 14:00 (11x), 93/E07: Do. 14:00 - 16:00 (13x), 93/E12: Do. 14:00 - 16:00 (11x), 93/E02: Do. 14:00 - 16:00 (11x), 94/E03: Dienstag, 27.01.2026 10:00 - 12:00, Dienstag, 27.01.2026 12:00 - 14:00, Dienstag, 10.02.2026 10:00 - 12:00, 94/E06: Dienstag, 27.01.2026, Dienstag, 10.02.2026 10:00 - 12:00, 94/E08: Dienstag, 27.01.2026, Dienstag, 10.02.2026 10:00 - 12:00, 94/E01: Dienstag, 10.02.2026 10:00 - 12:00

Zeiten: Di. 10:00 - 12:00 (wöchentlich), Ort: 66/E34, Do. 12:00 - 14:00 (wöchentlich), Ort: 93/E33, Do. 14:00 - 16:00 (wöchentlich), Ort: 93/E07, Do. 14:00 - 16:00 (wöchentlich), Ort: 93/E12, Do. 14:00 - 16:00 (wöchentlich), Ort: 93/E02, Termine am Dienstag, 27.01.2026 10:00 - 12:00, Dienstag, 27.01.2026 12:00 - 14:00, Dienstag, 10.02.2026 10:00 - 12:00, Ort: 94/E06, 94/E08, 94/E01

Beschreibung: Course Description:

Machine learning and neuroscience have a long intertwined history of trying to create and understand phenomena of intelligent, adaptive behaviour. The underlying connectionist approach has been highly influential for both fields. For machine learning, it led to the development of deep neural networks, which are built upon multiple biological inspirations (e.g. distributed coding, activation functions, stochasticity, dropout, attention, convolutions, etc). For computational neuroscience, neural networks act as a modelling framework for testing hypotheses of how distributed sets of simpler units can give rise to complex behaviour. In a true interdisciplinary fashion, these modelling efforts have benefited greatly from recent deep learning developments, as they allow for end-to-end trained systems performing complex tasks on real-world data. What emerged from this joint endeavour is the field of neuroconnectionism, which aims to integrate biological details into deep learning systems (architecture, learning objectives, and input statistics), while testing the resulting models against high-dimensional neural data and behaviour.
This course will introduce you to neuroconnectionism, i.e. how machine learning is used to make progress in modelling and understanding cognition.

Learning Objectives:

At the end of this course,
• you will have acquired knowledge in the emerging field of neuroconnectionism (i.e. deep learning for cognitive computational neuroscience).
• you will have learned about the use and caveats of computational modeling in general, and how (deep) neural networks are used for modeling cortical/cognitive function.
• you will have learned about different ways in which we try to reverse engineer the brain to (a) improve our understanding of neural mechanisms, and (b) to improve applications in artificial intelligence.
• topics will likely include biological and computer vision, reinforcement learning in man and machine, benefits of recurrent computations, prefrontal cortex and executive function, memory and replay, and continual, unsupervised and supervised learning.
• you can follow and discuss scientific talks by a domain expert in the area.
• you can organize, understand, and learn a large set of information and self-organize your learning individually, in teams, and in the class as a whole.
• you will have gained experience in creating a project on a state-of-the-art topic.

Prerequisites:

A basic understanding of neurobiology and neuropsychology (with a
special focus on vision), as well as a solid understanding of deep learning and linear
algebra

Course Format:

• Delivery Method: lectures: hybrid with recording // workgroup sessions: hybrid and in-person
• Type of Contact & Contact Hours: 2 hours a week lecture, 2 hours a week workgroup session
• Selection Process: instant entry

Assessment and Grading:

8 ECTS [200]
12 lectures (attendance, preparation and follow-up) [48]
10 workgroup sessions (attendance, preparation and follow-up) [30]
10 workgroup summaries + project work/report [84]
Self-study (background reading, exam study) [38]

1. Workgroup summaries (45 points)
Each week, you will submit a summary of max 200 words that will be due by the start of the workgroup the following week. This must be in paragraph form (no bullet-points) and will consist of two parts:

- A summary of the workgroup for that week (i.e. the video(s) you watched).
- A deeper exploration of an aspect of the topic you found most interesting. Here we ask that you choose a single topic from the video(s) and explore it further by doing some extra reading. Then you will describe that topic, citing at least one paper that does not come from the class.

Each workgroup summary counts for 5 points in total, 2.5 from the summary and 2.5 from the topic exploration.

2. Written exam (30 points)
Multiple choice

3. Project proposal (25 points)
You will be asked to propose your own research project on a topic related to the topics covered in the course. Such proposals are vital in academic life and are used in everything from thesis pitches to funding applications. The general goal is to show that you understand the background literature in the area you want to conduct a project in and have developed an appropriate project to tackle a question within that area.

You are not allowed to simply write a proposal for an existing scientific work, such as any of the studies covered in the course. The goal is to give you experience in developing your own independent research ideas.

Basic Guidelines:
800 words minimum, 1000 words maximum
At least 5 properly cited references (in APA format)

Important Dates:

• Final Exam: final lecture slot (27.01.2026)
• Application Deadline for the Exam (HISinOne/EXA): one week before the final exam (20.01.2026)
• Project proposal: 03.03.2026

Required Texts and Materials or further Resources:

As no textbook exists yet for the topics covered, the course is based entirely on peer-reviewed research articles. The PDF versions of them are shared via studip.

Will this class be offered again/regularly?

Yes, every winter term.

zur Veranstaltung in Stud.IP