Deep Learning
Syllabus, Master's level, 1RT720
- Code
- 1RT720
- Education cycle
- Second cycle
- Main field(s) of study and in-depth level
- Computer Science A1F, Data Science A1F, Image Analysis and Machine Learning A1F, Technology A1F
- Grading system
- Fail (U), Pass (3), Pass with credit (4), Pass with distinction (5)
- Finalised by
- The Faculty Board of Science and Technology, 29 February 2024
- Responsible department
- Department of Information Technology
Entry requirements
120 credits. Participation in Statistical Machine Learning and a second programming course. Proficiency in English equivalent to the Swedish upper secondary course English 6.
Learning outcomes
After passing the course the student should be able to:
- describe and use backpropagation together with gradient descent and stochastic gradient descent to optimize a model;
- implement a fully interconnected multilayer neural network from scratch;
- explain under- and over-fitting and what can be done to avoid these;
- compare and analyze different kinds of regularization techniques for deep neural networks;
- use modern environments for deep machine learning to solve practical data processing and analysis problems;
- construct and analyze deep neural networks for image data, time series data, and graph data;
Content
Data-driven regression and classification, linear classification, activation functions, various cost functions, gradient-based optimization with backpropagation. Convolutional neural networks (CNNs). Transformers. Graph neural networks (GNNs). Large-scale optimization and regularisation techniques for training deep models. Different architectures and applications. Possibilities and limitations with deep learning.
Instruction
Lectures, laboratory work, assignments.
Assessment
Written exam, oral and written presentation of assignments.
Other directives
Cannot be included in the same degree as 1MD120 Deep Learning for Image Analysis.
Reading list
No reading list found.