On this page, you can track the course's development over time. Once the course analysis is published, data for each course offering is displayed, including the number of registered students, course results, and planned improvements for the next session. All course syllabuses and course memos are shown on the page Archive.
The information can help prospective, current, and former students with course selection, or to follow up on their own participation. Teachers, course coordinators, examiners, program directors, and others can use the page as a resource for course development.
2026
Course analysis not published
2025
mladv25 2025-50325 ( Start date 27 Oct 2025, English )
Changes planned for the next course offering
Analysis
The course is viewed as relevant and engaging, with particularly strong results regarding inclusion, respect, and student participation. This indicates a well-functioning classroom environment and motivated students. The course will next year have a stronger focus on the probabilistic method and generative AI, which is likely to increase student satisfaction.
Areas that require attention concern clarity of course organization, transparency of expectations, and support for monitoring progress. The last component will be strengthen next year by a new examination system. This year fear of AI based cheating forced us to have substantial amounts of oral examination that removed resources from grading written solutions and, thereby, increased turn-around time. The variation in responses suggests that some students, especially those with weaker background knowledge, may have experienced uncertainty regarding structure, grading criteria, or alignment between teaching activities and examination.
Planned improvements
Further clarify course structure and expectations
Provide clearer grading criteria and rubrics for assignments.
Increase survey participation by reminding students and allocating time for evaluation completion.
The workload varies during the course. This will be countered by restructuring the assignments and also providing both assignments and projects at an earlier stage of the course.
The students desire to have a larger degree of collaborative moments. We will attempt to identify more implementation oriented parts of the examination that can be given as collaborative tasks.
In some videos the text have been smaller than what is desirable. These videos have been remade and lecture hall teaching will be used to introduce the material as well as follow up on the students understanding of it.
The examination has been based on the thresholding method recommended in some of the pedagogical courses at KTH, i.e., only those students with a grade above a certain threshold can proceed to the harder parts of the examination. This has led to frustration and stress among the students. We will simply abandon this system and allow all student to attempt all parts of the examination.
Many students start because of the hype of machine learning but do not finish the course and these may be diluting our Learning Experience Questionnaire. We will try to separate the two groups.