News
Presentation Schedule (and Reminder)Written on 21.11.25 (last change on 21.11.25) by Sebastian Stich Dear students, This is the (final) reminder to please register on LSF, if you have not yet done so. Thank you to everyone who has already registered for a presentation slot. If you haven't signed up yet, please do so as soon as possible. Based on the registrations so far, we can confirm that… Read more Dear students, This is the (final) reminder to please register on LSF, if you have not yet done so. Thank you to everyone who has already registered for a presentation slot. If you haven't signed up yet, please do so as soon as possible. Based on the registrations so far, we can confirm that presentations will be scheduled on:
(Friday, 4pm - 6pm) Please mark these dates in your calendar. Depending on further registrations, an additional slot may be added.
|
Overleaf and Presentation SchedulingWritten on 09.11.25 (last change on 09.11.25) by Sebastian Stich Dear students, I have shared an Overleaf document with each of you. Please check your spam folder if you did not receive the invitation, and let me know if you are still unable to access it. The document contains a LaTeX template for the final report and suggested weekly work packages. The… Read more Dear students, I have shared an Overleaf document with each of you. Please check your spam folder if you did not receive the invitation, and let me know if you are still unable to access it. The document contains a LaTeX template for the final report and suggested weekly work packages. The weekly tasks are not mandatory, but they are strongly recommended to help you make steady progress during the semester, rather than rushing at the end. Please verify that you understand your assigned paper. If you would like to swap topics, let me know by the end of next week and include a suggestion for which paper you would prefer. Please also select a presentation date in the shared Google document. We will likely need only 3–4 meetings for all presentations, so try not to spread them too far apart.
|
|
Written on 03.11.25 by Sebastian Stich Dear students, Due to scheduling conflicts, Friday at 4 p.m. is the only time that works for all participants. (No meeting this week; the schedule will follow later in the… Read more Dear students, Due to scheduling conflicts, Friday at 4 p.m. is the only time that works for all participants. (No meeting this week; the schedule will follow later in the semester.) Please register on LSF by November 24, 2025 to receive course credit. For now, please check your assigned topic and the corresponding papers. If you would like to swap topics or papers, let me know soon so that adjustments can be made. Further details about the organization will be shared later this week. |
Paper ListWritten on 27.10.25 by Sebastian Stich Dear students, I have added the link for the paper list (and voting) under the 'materials' tab. Please take a look a the list and indicate your preferences by Nov 1. Please provide information about any schedule conflicts for the seminar sessions this semester. |
Info SessionWritten on 23.10.25 by Sebastian Stich There will be a Zoom information session on Monday, October 27, at 4:00 PM, where I will briefly discuss the logistics of the seminar. A recording of this session will be made available for those who cannot attend. |
Seminar: Topics in Optimization for Machine Learning
Optimization lies at the heart of many machine learning algorithms. This seminar teaches an overview of modern mathematical optimization methods for machine learning and data science applications. In particular, we will discuss the theoretical basics of stochastic optimization, the scalability of algorithms to large datasets, and challenges in distributed optimization, such as federated machine learning and privacy aspects in optimization. The seminar is based on a mix of foundational papers and recent publications.
Previous attendance at the summer term lecture "Optimization for Machine Learning" or "Lectures on Modern Optimization Methods" is recommended but not required. A solid understanding of basic optimization principles is sufficient.
Organization
In this seminar, students will learn to present, discuss, and summarize papers in different areas of optimization for machine learning, and critically engage with peer work through reviews and discussion. Specifically, each student will get a topic assigned to them, consisting of two papers (a lead and follow-up paper). Each student will
- Write a short paper (max 6 pages) on the assigned topic, using the two provided papers as a starting point.
- Deliver a 20-minute presentation (plus 10 minutes for discussion).
- Write two short reviews (max 1 page each) of papers on other topics and prepare three discussion questions for each. The reviews will be shared with the group.
The final grade will depend on the presentation, the written deliverables, and the active participation in the discussions.
Important Dates
- Kick-off meeting 27.10.2025 (to be held online, via zoom).
- Registration on LSF: 24.11.2025
- The reviews (and questions) must be submitted during the semester.
- The presentations will be scheduled between December - February (grouped into clusters of talks)
- Hand-in of report: TBA.
Deliverables
2 short reviews (each contributes 10% of your final grade):Each review (max 1 page) should address:- What is the problem addressed by the paper?
- What was done before, and how does the paper improve on previous work?
- What are the strengths and the limitations of the techniques in the paper
- What part of the paper was difficult to understand?
- What are possible improvements or extensions of the techniques in the paper?
-
Participation in discussion (20%): Contribute to the discussion during the seminar meetings. To engage discussion, submit 3 written questions for each session the day before.
-
Presentation (60%): You will prepare and deliver a 20-minute presentation (followed by 10 minutes of questions/discussion) of the paper assigned to you. You will have the possibility to get feedback on your slides before the presentation.
-
Seminar Paper: (20%) You will write a seminar paper on the topic that you have presented. It must not be longer than 6 pages, not counting references and appendices. Papers can be shorter, but in general, the provided page limit is a good indicator of how long a paper should be.