Optimization for Machine Learning

Dear students,

We fixed the exam date to October 8. (For students who cannot attend on October 8, we offer an extra slot on October 6).

The mandatory exam registration on LSF has to be completed until 29 September 2026. (Please wait to register until you get confirmation that you "passed" the project).

If you already have other commitments on October 8 (or if you prefer to attend on October 6), please indicate this on the sheet (you find the link under 'materials').

 

Dear students,

For the project submission, please register your group as a team on the CMS website, ensuring that all team members have joined. Please then submit your work using the provided submission links. This ensures that our feedback will be visible to all team members.

If you encounter technical difficulties, it is also acceptable for one group member to submit the work on behalf of the entire group. If this also fails, you may send your submission to Parham and me by email. Please note that we cannot guarantee technical support late in the evening.

The project presentations are planned for Friday at 1:15 p.m. There will be no exercise session on Friday.

We currently plan to hold the exam on October 8, with October 6 as an alternative for students who cannot attend on October 8. We are still waiting for confirmation from the study office, so these dates are not yet final. Since many of you have asked about the exam date, I wanted to share our current plan. Further details will follow during the lecture.

Dear students,

A short reminder about Lectures 1 and 2, which are offered virtually.

Please use Monday and Tuesday to go through the slides and video recordings for both lectures and work through the corresponding exercises on the exercise sheet.

Afterward, please also take the online self-assessment quiz that we have uploaded. I recommend allowing around 45 minutes if you want to work through the questions carefully. The quiz should help you assess whether you are well prepared for the in-person part of the course or whether it would be useful to review some of the material again. You can take the quiz as many times as you like.

Our first in-person lecture will be on Wednesday, August 26, at 9:15am. We will briefly go through the course modalities at the beginning of the lecture. At the end of the lecture, we will also offer a short Q&A session on Lectures 1 and 2, so please bring any questions you may have about the material.

Lectures will be held in building E9 1 (CISPA), room 0.01 (presentation room, ground floor).

The first exercise session will be on Wednesday afternoon.

Best,
-Sebastian

Dear students,

Welcome to the summer block course on Optimization for Machine Learning.

The course will take place over the two-week period from August 24 to September 4, followed by the exam at a later date. On a typical course day, we will have a lecture in the morning, followed by an exercise session in the afternoon, with additional time for solving exercises and working on the mini-project. Since the course is very condensed, it is important that you set aside enough time over these two weeks.

I will be traveling on August 24 and 25, so the first two lectures will be offered asynchronously. I have already uploaded the slides and video recordings, and we will also upload the corresponding exercises.

In principle, you can work through these two lectures on August 24 and 25, but you are also welcome to start earlier. In particular, if concepts such as convex sets or gradient descent are new to you, I strongly recommend going through the material in advance so that everyone starts the in-person part of the course with a similar background.

The first in-person lecture will therefore take place on August 26, followed by the afternoon exercise session (we will also explain the course modalities in more detail during this first session).