Messages

The exam will be a standard written school exam (see the previous message for an example). You will be allowed to bring an approved calculator, but no other support material. In terms of the material covered by the exam, you will find a syllabus here. If you have any questions about the exam, or any other course-related questions, don't hesitate to ask.

Today's teaching session was the last one of the course. Now it is just a matter of preparing for the exam. For the purpose of practicing, you can find last year's exam here and a solution proposal here.

The mandatory assignment is now available here. The report should be submitted through Canvas and the deadline for submission is April 27th. Note that there is only one attempt(!) to pass the assignment and you need to make a genuine attempt to solve all the problems in order to pass (but everything does not have to be entirely correct). Please don't hesitate to ask if you have any questions about the problems or just want to discuss some of your proposed solutions before you submit. Happy solving!

There will be no teaching session on Tuesday after Easter (11.4). Instead, you will be given a bit more time to work on the problems in exercise sets 10 & 11, which are now available in the exercise page. We will then go through the problems together in the exercise class on Thursday (13.4). I will also hand out the mandatory assignment during that week. Happy Easter!

Due to illness, there will be no lecture tomorrow Tuesday (28.3), but the lecture will be postponed to Thursday (30.3).

There will be no teaching next week's Tuesday (28.2). Due to this, we will exceptionally have the lecture on Thursday (2.3) and go through the problems in exercise set 6 in the following week's exercise class (9.3). 

I have now uploaded solutions and an R script for the first exercise set to the exercise page. In terms of the solutions, I will upload detailed solutions to any problem I don't go through in class and shorter and sometimes partial solutions (or hints) to the problems I go through in class in detail. Please let me know if you need more details for some specific problem.

For those of you that could not attend the first lecture, the plan going forward is to have lectures on Tuesdays (10-12) and exercise classes on Thursdays (12-14). You will find the problem sets for each week under "Resources -> Weekly exercises", I will add the new problem set in the beginning of each week. In terms of format of the exercise class, the first hour (12-13) will be devoted to problem solving, that is, you get an additional hour of working on the problems (alone or together), and also the possibility to ask me any questions about the problems and/or course material. Then, in the second hour (13-14), we will go through the solutions together.    

The topic of the Spring 2023 version of STK4290/9290 is Probabilistic Graphical Models (PGMs). The course will give an introduction to the PGM framework, which aims at modelling a system of variables that interact with each other. The field of PGMs lie at the intersection of statistics and computer science, combining concepts from probability theory, graph algorithms and machine learning. We will look at the two most basic PGM representations: Bayesian Networks and Markov networks, and cover the main topics related to representation, inference, and learning.

As the main text book for the course, we will use: Koller, D. and Friedman, N. (2009). Probabilistic Graphical Models: Principles and Techniques. MIT Press. ISBN-13: 978-0262013192, ISBN-10: 0262013193.

Please send me an email if you have any questions about the course.