Pedagogical seminar "Learning with NOTEBOOKS"

Welcome to the next pedagogical seminar in the seminar series about the small GRU development projects that we started in 2018.

Time: Thu 2019-05-02 12.15 - 13.00

Lecturer: Associate Professor Arvind Kumar, EECS, Division of Computational Science and Technology

Location: 1) At Campus Valhallavägen: CS Biblioteket, Lindstedtsvägen 3, Floor 4. 2) At KTH Kista: Amiga

Modern neuroscience involves not only biological details of the structure of the brain and cognitive aspects of the brain function but also involves mathematical modelling and data analysis. Thus, Neuroscience [DD2401/3401] is one of the most interdisciplinary courses. The interdisciplinary nature of the course poses a big challenge to both teachers and students. The problem is exacerbated by the fact that there is no single textbook that covers all the aspects of neuroscience. While we can focus on a small subset of topics but that defeats the purpose of our course. We aim to provide a broad overview of the field to our students who take this course either out of curiosity or want to learn about challenging research topics for their doctoral studies. In my talk I will present how we are improving the teaching material of this course and learning experience of the students by using JUPYTER notebooks.

In general there is a lack of good teaching material on computational/mathematical modelling in neuroscience that is equally accessible to students from Engineering Physics, Computer Science and Biomedical Engineering. And it is no surprise that students find these topics the most difficult. Therefore, we have now created ‘Jupyter Notebooks’ for the topics in the course that involve mathematical modelling.

‘Jupyter Notebooks’ is a single document that allows the user to mix figures, mathematical equations, executable code, visualization and narrative text and allows us to present the topic as an interactive story. In the past year we have developed notebooks for essentially all the introductory topics covered in course DD2401 and even more advanced topics. The notebooks contain biological background, mathematical equations of the model and executable code. All the material is developed using opensource material. Students are required to use the code to perform simple experiments with the model and report their observations. Finally, they are required to write their own code or modify the existing code to perform some advanced exercises. Thus, the material is accessible to all students as working through the notebooks does not require strong mathematical background or programming skills.

A key feature of our notebooks is that students learn in a discovery mode. They perform simple experiments, make observations and are asked to develop hypotheses on specific questions. These hypotheses can be tested directly by modifying the code. Thus far the student experience has been very encouraging and students have even come forward to further develop the notebooks.​

See also this page (including slide sets of the previous talks)

NOTE! You do not need to sign up for the seminar. If you prefer, you can take your lunch with you.