WASP Deep Learning 2025

Objective 

The purpose of this course is to give a thorough introduction to deep learning, also known as deep neural networks. Over the last few years, deep machine learning has dramatically changed the state-of-the-art performance in various fields including speech recognition, computer vision, and reinforcement learning (used, e.g., to learn how to play Go). We focus initially on basic principles regarding how these networks are constructed and trained, we then move on to study optimization methods for training and finally, we cover a range of more advanced topics including generative adversarial networks (GANs). The overall objective is to provide a solid understanding of how and why deep machine learning is useful, as well as the skills to apply them to solve problems of practical importance.

Course goals

In the first course module, we aim to ensure that all students understand the basic concepts and tools in deep learning. The second course module will focus on training algorithms. The third course module contains research-oriented topics, knowledge of which will be useful in various PhD projects within WASP. This module contains three different topics.

Organization

The course is offered in spring 2025 and organized into three different modules:

• Module 1: 26-27 March, Chalmers
• Module 2: 23-24 April, Lund University
• Module 3: 21-22 May, KTH

Note: Please fill out this form related to Module 1 as soon as possible. 

Canvas access

If you have registered for the course, you should get a Canvas account in the coming days. The course page and most material are open for all, but to hand in assignments and take quizzes, you need to be able to log in.

If you are an external user (non-KTH), use this URL to log in: https://kth.instructure.com/login/canvas. If you don't know your password, you can get a new one by clicking "Glömt Lösenord" on that page. 

Course module 1: Deep Learning

March 26-27, Chalmers
Lennart Svensson

Note that you can find all the material related to module 1 under Modules. In particular, a detailed description of the first module and what is expected from you is available in Module 1.

Course module 2: Training Algorithm Fundamentals

April 23-24, Lund University
Pontus Giselsson

Information about Module 2 is available here.

Course module 3: Advanced Topics of Deep Learning

Hossein Azizpour

There are two options to pass module 3:

Option 1: Group project: information pdf  

Option 2: Two-day in-person meeting including quizzes and assignments on Nov 13 13:00 – Nov 14 15:00 (Place: Salongen room, in KTH Main Campus Library.) 

- Materials for in-person meeting: Module 3 - In-person Meeting Materials

Examination

To pass the course, all modules have to be completed.