Messages

Good job on the exam everyone! The grading process is ongoing and will be finished soon.

Also, I would greatly appreciate any feedback from you on the course concerning e.g. the lectures, slides, exercises, book, exam, etc. This is still only the second time the course was given and your comments/suggestions would help a lot when preparing the next version. You can send your comments by email to the student representative or directly to me. 

Thank you in advance and have a nice and relaxing summer!

There will be a digital lecturer's round (trøsterunde) over Zoom at 10:00-10:30 during the exam. I will add a link to the Zoom room in a message (like this one) here on the semester page. You just enter the waiting room via the link and I will admit you one by one. Alternatively, you can also contact me via email and I will try to answer as quickly as possible.

Also, if there turns out to be some ambiguities in the exam text that need clarification, I will add this information to the same message that contains the Zoom link. So, keep an eye on the message board and try hitting the refresh button a few times during the exam.

In general, the following guidelines apply in courses at the Department of Mathematics: 
- The examination lasts 4 hours. In addition, you will have an extra 30 minutes to scan and upload your PDF. 
- All examination aids are allowed (e.g. books, online resources, scientific programming tools, etc.). 
- It is not allowed to collaborate or communicate with others during the exam about the assignments. 

- You can be picked out for a meeting to check that you understand the content of your examination attempt. The meeting will not affect your grade, but it can result in a formal investigation into possible cheating. Read more about what is considered cheating on the UiO web pages: https://www.uio.no/english/studies/examinations/cheating/index.html

- The general guid...

To prepare for the exam, we will in the final exercise take a look at the exam from last year. We will go through the solutions two weeks from now on the Tuesday session (May 24th). Also, in the remaining three Wednesday time slots (May 11th, 18th and 25th), I will have a digital helpdesk type of thing where I will keep my Zoom room open and you can freely join to ask questions about the material covered in the course (you will find a Zoom link in the schedule). You can of course also send your questions over email if you prefer that.

We concluded the lectures today and final curriculum can now be found here. We don't have any exercise session tomorrow (4.5), but we will instead take a look at the solutions to the two final exercise sets (14 & 15) next week's Tuesday (10.5).

Note that the remaining teaching sessions in the course will be over Zoom. Next week we will finish the lectures, covering the rest of the causal inference part and also take a look at causal learning. There will be no exercise session next week, but instead we will go through the last exercise sets (14 & 15) the week after on Tuesday, giving you more time to work on the problems. As a final "exercise" before the exam, I will also give out the exam set from last year for you to practice on.

The mandatory assignment is now available. The report should be submitted through Canvas and the deadline for submission is April 21st (Thursday after Easter). Note that there is now only one attempt to pass the assignment. Please don't hesitate to ask if you have any questions about the problems. Happy solving!

There was a wish for more exercises, and I have now added a list of additional book exercises that have not been covered in the weekly exercises, yet are relevant to the material covered so far in the course. The purpose of the additional exercises is primarily for self-study, but I will try to upload (partial) solution proposals in due time also to these so you have something to check your solutions against. 

We will have a written home exam in this course, meaning that you will be allowed to use aids in the form of lecture slides, course book, internet, etc. More information about the exam will be provided later. Regarding the mandatory assignment, I will hand it out towards the end of March. The assignment will be similar to the weekly exercises in terms of format, a few paper-and-pen exercises (primarily non-* exercises) and one exercise that involves some coding.

We will change to physical teaching next week. However, since Aud 4 in the VB building is still not in use on Tuesday (Feb 8), we will have the lecture in the NHA building in room UE26 (one floor down from the lobby). Aud 2 & 4 should be back in use from Wednesday (Feb 9). I will take video recordings for those not being able to attend physically.

The VB building is temporarily closed, since the ventilation is being cleaned after a fire in the basement. We will therefore continue with Zoom teaching also next week.

You will find links to my solution proposals to the weekly exercises (and R scripts) on the exercise page. Let me know if you have any questions about the solutions.

I have now added today's lecture slides to its page (see Resources). From next week, I will add the lecture slides and exercise set for each week already on Monday.

Due to the current regulations, the two first weeks of the course will be taught digitally over Zoom. You will find a link to the Zoom room in the schedule. The first lecture is on January 18, and then I will give a brief introduction to the course and go through some basic introductory material related to probabilities and graphs.

The topic of the Spring 2022 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.