“Machine Learning for Signal Processing” is a master’s course at the Eindhoven University of Technology. In the course, students not only get hands-on experience with coding and implementing deep neural networks but also delve into their theoretical underpinnings. With a special focus on how deep learning can be viewed as a signal processing technique, this course is especially well suited for electrical engineering students. At the end of the course, students will be able to reason both theoretically and practically about deep learning techniques, weigh their pros and cons, and will be well-poised to use deep learning in their own research.

Lectures

1. Introduction*
2. Linear models*
3. Nonlinear models and optimization
4. Regularization and CNN*
5. Generative models*
6. Deep Unfolding

*Lecures are given by me.

Online Material

1. Introduction to generative modeling
2. VAE
3. GAN
4. Normalizing Flows