Course

Deep Neural Networks (ELE680)

Facts

Course code ELE680

Credits (ECTS) 5

Semester tution start Autumn

Language of instruction English

Number of semesters 1

Exam semester Autumn

Time table View course schedule

Literature Search for literature in Leganto

Content

Note! This is an elective course and may be cancelled if fewer than 10 students are enrolled by August 20 for the autumn semester/January 20 for the spring semester. 

In this course, you will be introduced to the foundations of deep learning, basic network structures and their applications and how to build, train and evaluate deep neural networks for different applications. This includes:

  • Neurons, layers, back propagation, optimizers, loss functions, hyperparameters
  • Multilayer Perceptron Network (MPN)
  • Training a Neural Network
  • Unsupervised, supervised and semi-supervised learning approaches.
  • Transfer learning
  • Convolutional Neural Network (CNN)
  • Recurrent Neural Network (RNN) and Long Short-Term Memory networks (LSTMs)
  • Computer vision applications
  • Transformers
  • Text and Natural Language Processing

Learning outcome

At the end of this course the student should have a fundamental understanding of methods and deep neural network structures commonly used in deep learning. The student should also be able to build, train and evaluate models for one or more practical deep learning problems.

Required prerequisite knowledge

None

Recommended prerequisites

Machine Learning (ELE520)

Exam

Report and oral exam

Weight 1/1

Marks Letter grades

Report

Weight 3/5

Duration 1 Months

Marks Letter grades

Oral exam

Weight 2/5

Duration 10 Minutes

Marks Letter grades

The course has a continuous assessment consisting of a report and an oral exam. Both assessment components must be passed within the same semester to achieve an overall grade in the course.

Students who wish to improve their grade in the course must complete both assessment components again the next time the course is taught.

The report is based on a project carried out in groups.

The oral exam is individual and will have a certain connection to the project. The timing is adapted as needed, but no retake exam is offered.

Coursework requirements

2 assignments
2 out of 2 exercises must be approved within the specified deadlines. Approved exercises are valid for one year.

Method of work

The course has a duration of approximately 12 weeks and will be completed early November. Lectures will be held the first 7 weeks. The students are expected to spend additional 5 hours a week on self-study and assignments.

The project will be carried out in the last five weeks of the course, and it is expected that each student spends about 10 hours per week.

Open for

Admission to Single Courses at Master Level at the Faculty of Science and Technology
Data Science - Master Computational Engineering - Master Computer Science - Master Computer Science - Master (Part-Time) Cybernetics and Applied AI - Master
Exchange programme at The Faculty of Science and Technology

Admission requirements

Must meet the admission requirements of one of the study programmes the course is open for.

Course assessment

The faculty decides whether early dialogue will be held in all courses or in selected groups of courses. The aim is to collect student feedback for improvements during the semester. In addition, a digital course evaluation must be conducted at least every three years to gather students’ experiences.

Contact

Head of Department:
Tom Ryen
Course teacher:
Øyvind Meinich-Bache
Course coordinator:
Øyvind Meinich-Bache

Offered by

Faculty of Science and Technology

Department of Electrical Engineering and Computer Science

The course description is retrieved from FS (Felles studentsystem). Version 1