Fall 2024 - Contemporary Machine Learning and Nonparametric Methods

September 2024 - December 2024, University of St. Gallen

The Method Reading Group takes place bi-weekly with sessions of 1h15m, and we aim to devote at least one hour to topic discussion.

This semester, we meet on Thursdays, 16:15-17:30. Please see the schedule section below for updated room locations, dates and topics.

The group is open to anyone: if you would like to participate in the discussion too, let us know!

• Contact:
  • Giovanni Ballarin (giovanni.ballarin[-at-]unisg.ch)
  • Erik-Jan Senn (erik-jan.senn[-at-]unisg.ch)
• Files:
  • See below for materials discussed
  • Some additional references are also listed

Locations

Main Room:

Room C 83-1235, Haus Washington
Rosenbergstrasse 20/22, St. Gallen
Link to MazeMap

At the main foyer of the building, take the right door (coming from the street) towards the decorated staircase.
The seminar room is on the 1st floor (just above the steps), in front of the first elevator.

Other:

• Room C 61-152, Computer Science Building - [only 26.09]
  Rosenbergstrasse 30, St. Gallen
  Link to MazeMap

Schedule

Please check this website for regular updates

Materials

• Session 1
  • Annotated Paper
  • Handwritten Notes
  • See Chapters 2-6 & 11 of Zhang (2023)
  • See Chapters 4 & 9 of Bach (2024)
• Session 2
  • Annotated Paper
  • Handwritten Notes
• Session 3
  • Annotated Paper
  • Handwritten Notes
• Session 4
  • Annotated Paper
• Session 5
  • Annotated Chapter

Additional References

1. Zhang - Mathematical Analysis of Machine Learning Algorithms (2023)
   This recent book covers much of the methods (and the theory needed to study them formally) of contemporary Machine/statistical Learning, including kernel methods, additive models and neural networks. It also discusses online learning.

2. Bach - Learning Theory from First Principles (2024)
   Similar to the previous ref., this book also presents nonparametric and ML methods, but it also starts with an in-depth discussion of linear models. Additional topics include ensemble learning, overparametrized models and optimization.

3. Shalev-Shwartz & Ben-David - Understanding Machine Learning (2014)
   General book that covers (not in detail) the wide landscape of popular ML approaches and algorithms.

4. Tsybakov - Introduction to Nonparametric Estimation (2009)
   Modern (but already classic) reference for the foundational theory of kernel/series methods and minimax analysis.

Members

• Giovanni Ballarin
• Prof. Matthias Fengler
• Luca Fiumana
• Prof. Lyudmila Grigoryeva
• Christoph Hirt
• Jonas Huwyler
• Prof. Jana Mareckova
• Erik-Jan Senn

How To

The general guidelines and "house rules" we follow are much inspired by those of e.g. the TS&ML Reading Group at the University of Southampton.

We will alternate over time so that each person can try and fulfill the two main roles at least once:

• Presenter: the person who presents (with annotated paper, whiteboard, slides...) the meeting's topic -- they act as the 'key expert' on the chosen paper/chapter/topic, and go over the core points of the material, step by step. The presenter should be well-prepared in the content presented, and be able to provide a compact but transparent and technically accessible presentation of the material.

• Discussant: the person that, having read in some detail the material pertaining to the the meeting, can help steer the discussion: what are some key issues or questions? What things may be unclear? What are the strengths and weaknesses? Your role is to help everyone in the session navigate and understand the main ideas discussed.

The following (total) preparation times are suggested:

• Presenter: approx. 10 hours
• Discussant: approx. 5 hours
• General audience: approx. 2 hours

If you encounter any issues with the materials, do not hesitate to contact us!