Neural Differential Equations

Upon completion of the course, the student will be able to:
• utilize the methods used in discrete dynamical systems in analyzing the deep neural networks
• see the equivalence between deep neural networks and neural differential equations
• use the modern computational methods to solve neural differential equations
• utilize the operator learning techniques to reduce the curse of dimensionality in several real life problems

Advanced engineering mathematics

Keijo Ruotsalainen / Marko Huhtanen

Lecture notes and journal articles. Access to the Moodle requires registration for the implementation.

Attending the lectures, solve given assignments and final exam.

- Contact teaching
- Material will be in Moodle
- Lectures 30 h, self-study 80