Skip to main content
Till KTH:s startsida

DD2610 Deep Learning, advanced course 7.5 credits

Administer About course

Deep learning has dominated the field of AI and machine learning. Therefore one basic course cannot meaningfully cover all the relevant topics of deep learning research. This master-level course goes through several advanced topics of deep learning. Therefore, the course requires solid prior knowledge in machine learning in general and deep learning in particular. 

The topics are chosen based on two main factors: (1) primarily areas which are deemed most relevant for potential PhD studies in machine learning and for a continuation in the industry as an ML engineer. (2) not all relevant topics are possible to cover in a single course, therefore the relevant topics are narrowed down to those where the teachers have strongest research activities.

Currently, the included topics are: 

  • Probabilistic Deep Learning (uncertainty quantification with deep networks)
  • Deep Generative Modeling
  • Understanding deep learning and interpreting deep networks
  • Under-supervised Deep Learning
  • Deep Learning for Scientific Discovery
  • Best practices in machine learning research.

Note that there are several other advanced topics of deep learning which are not currently covered such as advanced architectures, geometric deep learning, deep reinforcement learning, among others.

 

The course spans two periods and contains the following items:

  • Period 1 Lectures:
    • pre-recorded video lectures on the fundamental topics to be watched at home
    • in-person lectures on recent topics 
    • discussion sessions on students' questions and comments
  • Period 2 Lectures
    • None
  • Period 1 Assignments
    • one formative quiz on each topic
    • one implementation practical on each topic
    • a form to fill in on each topic to share questions and comments
    • final project proposal
  • Period 2 Assignments
    • final project work and submission of the report.

Information per course offering

Termin

Information for Autumn 2026 Start 24 Aug 2026 programme students

Course location

KTH Campus

Duration
24 Aug 2026 - 11 Jan 2027
Periods

Autumn 2026: P2 (3 hp), P1 (4.5 hp)

Pace of study

25%

Application code

11586

Form of study

Normal Daytime

Language of instruction

English

Course memo
Course memo is not published
Number of places

Min: 1

Target group
Open for students from year 3 and students admitted to a master's programme as long as it can be included in your programme.
Planned modular schedule
[object Object]

Contact

Examiner
No information inserted
Course coordinator
No information inserted
Teachers
No information inserted

Course syllabus as PDF

Please note: all information from the Course syllabus is available on this page in an accessible format.

Course syllabus DD2610 (Autumn 2026–)
Headings with content from the Course syllabus DD2610 (Autumn 2026–) are denoted with an asterisk ( )

Content and learning outcomes

Course disposition

 

Course contents

  • Basic theory of deep networks
  • Probabilistic deep learning
  • Types of learning in deep learning
  • Deep learning with imperfect labels
  • Reliable deep learning

Intended learning outcomes

On completion of the course, the student should be able to

  • explain and justify the subareas of deep learning
  • account for the theoretical background for advanced deep-learning techniques
  • identify needs of additional research in the area
  • implement several deep-learning methods based on new research
  • analyse advanced research in the area and critically evaluate the methods' weaknesses and strengths

in order to:

  • be prepared for a degree projects/postgraduate studies in deep learning
  • better be able to meet the industry's need for cutting-edge competence in the area.

Literature and preparations

Specific prerequisites

Knowledge in basic deep learning, 7.5 credits, equivalent completed course DD2424/DD2437.

Active participation in DD2424 during study period 4 of the same calendar year is equated with completed course. Anyone who is registered is expected and considered to be actively participating.

Literature

You can find information about course literature either in the course memo for the course offering or in the course room in Canvas.

Examination and completion

Grading scale

A, B, C, D, E, FX, F

Examination

  • PRO1 - Project work, 3.0 credits, grading scale: A, B, C, D, E, FX, F
  • LAB1 - Laboratory work, 4.5 credits, grading scale: P, F

Based on recommendation from KTH’s coordinator for disabilities, the examiner will decide how to adapt an examination for students with documented disability.

The examiner may apply another examination format when re-examining individual students.

If the course is discontinued, students may request to be examined during the following two academic years.

Examiner

Profile picture Hossein Azizpour

Ethical approach

  • All members of a group are responsible for the group's work.
  • In any assessment, every student shall honestly disclose any help received and sources used.
  • In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.

Further information

Course room in Canvas

Registered students find further information about the implementation of the course in the course room in Canvas. A link to the course room can be found under the tab Studies in the Personal menu at the start of the course.

Offered by

EECS/Robotics, Perception and Learning

Main field of study

Computer Science and Engineering

Education cycle

Second cycle

Supplementary information

Overlaps with DD2412.

In this course, the EECS code of honor applies, see:
http://www.kth.se/en/eecs/utbildning/hederskodex