Advanced Topics in Deep Learning (ATDL)
Course content
This course provides a detailed insight into advanced deep learning methods, covering algorithms, theory, and tools in this rapidly evolving field.
As an advanced-level course, it builds on deep learning basics and therefore has machine learning and deep learning courses as prerequisites.
The focus will be on advanced deep generative models, graph neural networks, foundation models, and methods for efficient training and transfer learning.
The topics may vary from year to year, reflecting current trends
in literature.
MSc Programme in Computer Science
Knowledge of
Selected advanced topics in deep learning, including:
-
State-of-the-art deep models in specific domains
-
Design and analysis of deep learning algorithms
-
Theoretical foundations of deep learning
The exact list of topics will depend on the teachers and current
research trends in deep learning. Topics will be announced on the
course's Absalon page.
Skills to
-
Implement advanced deep learning algorithms using state-of-the-art tools
-
Read, understand, and assess recent scientific literature in deep learning
-
Apply knowledge obtained through research papers to practical and theoretical problems
-
Compare deep learning methods and evaluate their strengths and limitations
Competences to
-
Understand advanced deep learning methods and techniques
-
Design, optimize, and apply advanced deep learning models
-
Plan, structure, and carry out self-learning within the field
The course will mix lectures, exercise classes, and project work.
See Absalon.
Applicants are expected to have academic qualifications
equivalent to a BSc degree.
The following courses are highly recommended:
- Machine learning: corresponding to Machine Learning A (MLA) and
Deep Learning (DL)
- Linear algebra: corresponding to Linear Algebra in Computer
Science (LinAlgDat)
- Calculus: corresponding to Mathematical Analysis and Probability
Theory for Computer Scientists (MASD) and Modelling and Analysis of
Data (MAD), or Introduction to Mathematics for Science
(MatIntroNat) and Mathematical Analysis (MatAn)
- Statistics and probability theory: corresponding to Probability
Theory and Statistics (SS)
- Programming: solid experience in Python
- ECTS
- 7,5 ECTS
- Type of assessment
-
Continuous assessment
- Type of assessment details
- Continuous assessment of three written assignments and group presentations. All assignments must be passed. The final grade is based on an overall assessment of the assignments and presentations.
- Aid
- All aids allowed
- Marking scale
- 7-point grading scale
- Censorship form
- No external censorship
Internal examiner.
- Re-exam
-
The re-exam is a 25-minute oral examination, without preparation, in the course syllabus.
Criteria for exam assessment
See Learning outcome
Single subject courses (day)
- Category
- Hours
- Lectures
- 28
- Class Instruction
- 14
- Preparation
- 70
- Exercises
- 94
- English
- 206
Kursusinformation
- Language
- English
- Course number
- NDAK24003U
- ECTS
- 7,5 ECTS
- Programme level
- Full Degree Master
- Duration
-
1 block
- Placement
- Block 1
- Schedulegroup
-
A
- Capacity
- No limitation. Unless you register in the late-registration period (BSc and MSc) or as a credit or single-subject student.
- Studyboard
- Study Board of Mathematics and Computer Science
Contracting department
- Department of Computer Science
Contracting faculty
- Faculty of Science
Course Coordinator
- Mostafa Mehdipour Ghazi (5-6d6e67806f466a6f34717b346a71)
Er du BA- eller KA-studerende?
Kursusinformation for indskrevne studerende