IE 678: Deep Learning (FSS 2024)

Organization

Lecturer:Prof. Dr. Rainer Gemulla, Simon Forbat
Type of course: Lecture (in presence), exercises, assignments (6 ECTS points)
Prerequisites: IE 675b Machine Learning or equivalent
Registration: enroll in ILIAS

Lecture (Thursday) starts in the first week, tutorial (Tuesday) in the second week. Lecture notes, exercises, assignments, and supplementary material can be found in ILIAS.

Content

Machine learning is concerned with building computer systems that improve with experience as well as the study of learning processes, including the design of algorithms that are able to make predictions or extract knowledge from data. Building upon IE 675b Machine Learning, this course focuses on deep learning and introduces basic and advanced deep learning architectures and techniques, training methods and hyperparameter optimization, as well as selected applications.

Tentative topics include:

Feedforward neural networks
Backpropagation and parameter optimization
Machine learning systems
Training techniques for deep learning models
Recurrent neural networks
Convolutional neural networks
Attention and Transformers
Deep learning for graphs
Deep generative modelling

Literature

I. Goodfellow, Y. Bengio, A. Courville. Deep Learning. The MIT Press, 2017 (http://www.deeplearningbook.org/)
K.P. Murphy. Probabilistic Machine Learning: An Introduction. The MIT Press, 2022 (https://probml.github.io/pml-book/book1.html)
K.P. Murphy. Probabilistic Machine Learning: Advanced Topics. The MIT Press, 2023 (https://probml.github.io/pml-book/book2.html)
I. Drori. The Science of Deep Learning. Cambridge University Press, 2023 (eBook)
Additional material and articles provided in lecture notes