Electrical Engineering MA, Machine Learning on Embedded Systems, 7.5 credits

Aim

The course aims for students to develop an understanding of the opportunities and challenges of machine learning integration on resource-limited embedded systems, such as smart sensors and IoT devices. The course, moreover, aims to develop the students' skills to adapt and deploy machine learning models for integration on such embedded systems, as well as to analyse the model performance and effects of model optimisation.

Course objectives

After the successful completion of the course, the student should be able to:
- describe key terms and concepts of embedded machine learning, and compare the differences of embedded machine learning with more traditional machine learning implementations.
- exemplify applications that are suitable for embedded machine learning.
- apply state-of-the-art methods and use tools for embedded machine learning.
- summarise and analyse research results on the application of embedded machine learning.
- evaluate ethical consequences of performing machine learning on embedded systems.

Content

- Motivation, possibilities and needs for embedded machine learning.
- Applications of embedded machine learning.
- Differences between embedded machine learning and machine learning on more traditional computing systems.
- Resource-limitations of embedded systems and influences of hardware architectures.
- Software optimization of machine learning models for embedded systems.
- Tools and methods for embedded machine learning.
- Ethical aspects of machine learning.

Entry requirements

Electrical Engineering BA or Computer Engineering BA, 45 credits, including microcontroller programming and an introductory course in machine learning.

Selection rules and procedures

The selection process is in accordance with the Higher Education Ordinance and the local order of admission.

Teaching form

This course contains lectures, seminars and laboratory sessions. The course is normally given in English.

Examination form

I101: Literature study, Assignment Report, 3 Credits
Grade scale: Seven-grade scale, A, B, C, D, E, Fx and F. Fx and F represent fail levels.

I201: Methods and tools, Assignment Report, 3 Credits
Grade scale: Seven-grade scale, A, B, C, D, E, Fx and F. Fx and F represent fail levels.

Q101: Terms and concepts, Test/Quiz, 1.5 Credits
Grade scale: Fail (U) or Pass (G)

Grading criteria for the subject can be found at www.miun.se/gradingcriteria.

The examiner has the right to offer alternative examination arrangements to students who have been granted the right to special support by Mid Sweden University’s disabilities adviser.

Examination restrictions

Students are entitled to three examination opportunities within one year according to the examination format given in this version of the course syllabus. After the one-year period, the examination format given in the most recent version of the course syllabus applies.

Grading system

Seven-grade scale, A, B, C, D, E, Fx and F. Fx and F represent fail levels.