Projects at the intersection of neuroscience and machine learning

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

Veranstaltungstyp: Seminar

Ort: (50/E07): Mi. 10:00 - 11:30 (14x), (50/106): Mi. 14:00 - 16:00 (14x)

Zeiten: Mi. 10:00 - 11:30 (wöchentlich), Mi. 14:00 - 16:00 (wöchentlich)

Beschreibung: Course description:

In this course, you will work on your own in-depth project, either alone or (preferably) with at least one other student. The projects can be chosen from a list provided by the instructors or decided jointly with students and instructors. The course will begin with in-depth discussions to help you decide the details of the projects. In the subsequent weekly meetings, you will provide brief summaries of your progress - what is done, what are the roadblocks, and what are the next steps, and the group and instructors will provide guidance on how to proceed further. Students will be required to review each-other's code to learn to write clearly/accessibly, and to take the perspective of an external code-reviewer. By the end you will have completed a project at the intersection of neuroscience and machine learning, you will have learned to review code, to communicate problems, and you will have gotten an in-depth insight into the research field as such. You will be required to document your project in the form of a research paper or a thesis.

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 completed a research project at the intersection of neuroscience and machine learning
• you will have learned to review code, to communicate problems, and to get an in-depth insight into the research field
• you can follow and discuss scientific talks by a domain expert in the area.

Prerequisites:

Master students with the following experiences/coursework completed:
• Proficiency in programming with python
• Introduction to Linear Algebra
• Neuroscience basics (e.g. Action and Cognition (Vision))
• Implementing ANNs in Tensorflow or Machine Learning for Cognitive Computational Neuroscience

Course Format:

This is a work intensive course with two meetings each week:
• One meeting is to be an active part of the lab colloquium: to inform yourselves about other work in the area, to contextualise your work, and to present your plans and work.
• Another weekly meeting is for course-members only and will be used to discuss the project, code, progress, roadblocks, etc.

Number of participants:

The number of participants of this course is limited to ensure that we can offer the best in-depth project guidance and support. If more students than the maximum number apply, we will choose a subgroup based on fulfilment of course criteria/past experience and motivation letters. The exact procedure will be communicated in the first session once we can tell how many students exactly are actively participating.

Course application:

If you are interested in taking part in the course, please provide proof that you fulfil the course criteria (see above) as well as a short motivation letter (250 words max). We will provide you with a template for this before the first session of the course.

Assessment and Grading:

If you participate with a standalone project your grade will be based on your project, documentation and your active participation in the weekly meetings. If you take this seminar alongside your thesis work, no credit/grade can be given.

4 ECTS
Weekly lab colloquium attendance and participation; weekly project meeting attendance and participation; Project work and documentation; Self-study (background reading, project research).

Required Texts and Materials or further Resources:

None

Important Dates:

• Final Project Documentation Due: 03.03.2026
• Application Deadline for the Course: one week before the first session. Please sign up on Stud.IP to be informed about the application round.

Will this class be offered again/regularly?

Yes, every term.

zur Veranstaltung in Stud.IP