Teaching ML 2022

3 papers
The 2022 edition of Teaching ML focused on the intersection of machine learning theory and pedagogical practice. Key research directions included the study of benign overfitting in overparameterized models and the diagnosis of reasoning failures through fine-grained error taxonomies. Researchers also explored the empirical validation of modern pedagogical frameworks and the decoupling of logic from syntax to improve technical education. Notable work addressed how students perceive core concepts, as seen in the paper "Machine Learning Students Overfit to Overfitting," which examined student understanding of model generalization. Another prominent trend involved the use of multi-agent environments and strategic reasoning to facilitate learning. The "Hearts Gym" project demonstrated this by using team-based tournaments in imperfect information games to teach reinforcement learning. Additionally, the year’s contributions emphasized the development of modular, open-source educational resources for machine learning and data science. Prolific contributors included authors such as Matias Valdenegro-Toro, Bernd Bischl, and David Rügamer, who focused on creating flexible and scalable curricula to support diverse learning needs in the field.

Top Research Directions

The 8 research directions with Teaching ML 2022 papers, ranked by paper count, next to 2021. Share is the fraction of the edition’s papers in the direction.

#Direction20222021Share 2022Share 20212021 rank
1Competitive Round Robin Tournaments for Dynamic Agent Evaluation1033.3%0.0%—
2Decoupling Language Syntax from Computational Logic1033.3%0.0%—
3Fine-Grained Failure Mode Decomposition for Language Models1033.3%0.0%—
4Policy Invariant Reward Shaping1033.3%0.0%—
5Sample Efficient Reinforcement Learning1033.3%0.0%—
6Shift-Left Testing and Continuous Verification1033.3%0.0%—
7The Mechanics of Benign Overfitting in High Dimensions1033.3%0.0%—
8The Reward Engineering Bottleneck in Reinforcement Learning1033.3%0.0%—

Download all directions as CSV

Institutions

Institutions credited by Teaching ML 2022 papers: a paper credits every institution named in its affiliations, once. 7 institutions share 3 credited papers.

#InstitutionPapersShare20212021 rank
1011k1gy03 133.3%0—
2Forschungszentrum Jülich 133.3%0—
3Leibniz University Hannover 133.3%0—
4Ludwig-Maximilians-Universität München 133.3%0—
5Munich Center for Machine Learning 133.3%0—
6RWTH Aachen University 133.3%0—
7University of Groningen 133.3%0—

Download all institutions as CSV

About this data

Papers. Teaching ML 2022 lists 3 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers. Comparisons with 2021 use shares, since the edition went from 2 to 3 papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 3 of 3 papers (100.0%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-19 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “Teaching ML 2022”, https://lacuna.tiptreesystems.com/venue/teaching-ml/2022.

Top Authors

AuthorPapers
Bernd Bischl1
David Rügamer1
Giuseppe Casalicchio1
Jan Ebert1
Ludwig Bothmann1
Marius Lindauer1
Stefan Kesselheim1

Impactful Papers

Papers from Teaching ML 2022 that shaped research clusters.

Influential

Papers

Title
Developing Open Source Educational Resources for Machine Learning and Data Science
Hearts Gym: Learning Reinforcement Learning as a Team Event
Machine Learning Students Overfit to Overfitting