Høst 2021

DTE-2502 Neural Networks - 10 stp

The course is administrated by

Institutt for datateknologi og beregningsorienterte ingeniørfag

Type of course

The course can be taken as a single course.

Course contents

The course focus on different approaches into the artificial intelligence domain focusing on neural networks. The course is divided into a theoretical part and a practical exercise scheme where the candidates will work in teams to develop and use neural nets in a robotic setting. The candidates will work in groups of 3-4 individuals, assembling the robot, implement navigation and movement based on Kalman or another appropriate AI algorithm. The course focus on topics such as application / programming / understanding of concept of a perceptron, the multi-perceptron, feed-forward neural nets, support vector machines (SVM), recurrent neural networks (RNN), convolution neural nets (CNN) and a basic introduction to deep methods. Emphasis will be placed on applications related to image classification and various forms of regressions.

Objective of the course

On completion of the course, the successful student is expected to have the following:

Knowledge

The student will have:

An overview of history and numerous approaches within machine learning neural nets.
Understanding of "The curse of dimensionality" in AI.
Basic understanding of back-propagation and complexity.

Skills

The student should be able to:

Program, adapt and apply neural nets in different application domains.
Identify and define features in a complex environment.

General Competence

Can apply the knowledge and skills to solve problems and communicate about the results with other specialists in the field of computer science.

Language of instruction

English

Teaching methods

The subject uses so-called "Flipped classroom", i.e., lectures are posted online continuously during the semester in the form of short instructional videos and demonstrations of 10-20 minutes. In addition, exercises and control questions related to each video are used.

The subject teaches in the autumn semester with teacher-led and assistant-led learning and / or exercises. Online students will have access to a teaching assistant for afternoon / evening support.

Assessment

Course work requirement (work requirements)

Mandatory exercises: 4 of 6 approved exercises (Pass / Fail)

Examination and assessment:

Folder assessment with the following content (assessment basis):

Mandatory work: 2 programming works (Grade: A-F)
Multiple choice test for parts of the syllabus (Grade: A-F)

Work and multiple-choice tests can be weighted differently in their contribution to the final grade in the course.

If the work requirements are not met (delivered and passed at least 4 of 6 submissions) the candidate will not qualify to get a grade in the course.

If more than 30% of the portfolio (assessment basis) is missing (i.e., has passed less than 70% of the portfolio), the candidate will not qualify for continuation and must re-take the course at the next ordinary period.

The portfolio (exercises and tests) can be delivered in Norwegian or English.

Continuation

Continuation consists of completing the folder within a deadline. The parts of the work requirements that were missing during ordinary deadline will be replaced by new corresponding work requirements that must be made within the agreed deadline.

Date for examination

Portfolio hand in date 14.12.2021

The date for the exam can be changed. The final date will be announced at your faculty early in May and early in November.