Advanced Topics in Machine Learning

Registration deadline: 16 March 2026

Type of activity: PhD Reading group

Organizer(s):
Pınar Tözün, Associate Professor, 
Jaike van Twiller, Postdoc,
Robert Bayer, PhD Fellow,

Lecturer(s):
Pınar Tözün, Associate Professor,

Date(s) of the course:
Every Thursday
April 2, 2026 – June 18, 2026 (12 weeks)

Time:
15:00-16:00

Room:
TBD

Course description:
Machine Learning research is characterized by its high velocity and increasing specialization. This reading group provides a structured forum to bridge the gap between specialized research areas, ensuring that breakthroughs in one sub-field, such as optimization, machine learning systems, computer vision, or natural language processing, can inform and inspire work in others through active peer-to-peer exchange.

Intended learning outcomes are the following:

• Critically evaluate state-of-the-art research papers in various machine learning sub-fields.
• Present complex technical concepts to a diverse audience of peers from different research groups.
• Identify and articulate potential cross-disciplinary collaboration opportunities within the internal ML community at ITU.

Reading list:
Besides reading a research paper in preparation for every week, the students are expected to read the following mandatory reading, which will set the fundamentals of how to critically evaluate the research papers the students will read throughout the course.

S. Keshav. 2007. How to read a paper. SIGCOMM Comput. Commun. Rev. 37, 3 (July 2007), 83–84.

Sayash Kapoor et al. ,REFORMS: Consensus-based Recommendations for Machine-learning-based
Science.Sci. Adv.10,eadk3452(2024).

Programme:
Every week one of the participants picks the research paper, presenting the cutting-edge research in their research field. The students are required to present at least one and up to two research papers, based on the number of participants. The schedule and the research papers presented, will be set once the participants are chosen.

Assessment:
The students are assessed on pass/fail basis. To pass, a student must lead at least one session, presenting a research paper from their research area, while utilizing the framework from the
mandatory reading material, and actively participate in discussions on papers presented by other students.

Credits:
3 ECTS

Number of hours the student is expected to use on the course:
Participation: 1 hour / week (12 hours total)
Preparation: 6 hours / week (72 hours total) + 3 hours to prepare presentation during the student’s turn

How to sign up:
Students should sign up by emailing Robert Bayer (roba@itu.dk) by March 16, 2026.