16

Apr

Teaser image to Some Recent Advances in Exceptional Model Mining: Tales of Potatos, Boris Johnson, and Atrial Fibrillation

Lecture

Some Recent Advances in Exceptional Model Mining: Tales of Potatos, Boris Johnson, and Atrial Fibrillation

Wouter Duivesteijn, TU Eindhoven

   16.04.2024

   2:00 pm - 3:30 pm

   LMU Main Building, Room M001, Geschwister-Scholl-Platz 1, 80539 München

This MCML workshop covers the topic of Exceptional Model Mining (EMM). EMM finds exceptional subgroups in data by partitioning columns for candidate subgroup definition and evaluation of the exceptionality of subgroup behaviour.

Exceptional Model Mining (EMM) strives to find subgroups in a dataset that behave somehow exceptionally. We partition the columns of the dataset into two sets. The one set is used for defining candidate subgroups (by making conjunctions of conditions on individual attributes); a challenge lies in efficiently traversing the search space of candidate subgroups. The other set is used for evaluating exceptionality of subgroup behavior: we define a kind of interaction between these target columns, and deem subgroups to be exceptional if parameters of this interaction have exceptional values.

We will illustrate this concept with three types of interaction. The first, regression, has implications for the price of rice in China (traditional linear regression), dot counting skills of Finnish school children (piecewise regression), and potato growth (mixed models). The second, exceptional trends in repeated cross-sectional data, illustrates how local deviations can make Boris Johnson appear in a graph of trends. The third, exceptional signals in EGCs, has the potential to improve detection of atrial fibrillation.

If time remains, we will discuss how EMM relates to clustering; there are clear connections and clear differences between the tasks, but are there fruitful avenues for cross-pollination?

Organized by:

Munich Center for Machine Learning

Lehrstuhl für Datenbanksysteme und Data Mining LMU Munich


Related

Link to Foundational Methods for Foundation Models for Scientific Machine Learning

Munich AI Lectures  •  09.07.2025  •  Plenarsaal of the Bavarian Academy of Sciences and Humanities (BAdW), Alfons-Goppel-Straße 11, 80539 Munich

Foundational Methods for Foundation Models for Scientific Machine Learning

The Munich AI Lectures on March 26, 2025, will feature Michael Mahoney and cover foundational methods for SciML foundation models.


Link to The Best of CVPR 2025 Series

Meetup  •  09.07.2025  •  Online

The Best of CVPR 2025 Series

Voxel51’s “Best of CVPR” event features top 2025 computer vision research, including Cecilia Curreli presenting on human motion prediction.


Link to Festival of the Future 2025

Festival  •  03.07.2025 - 06.07.2025  •  Deutsches Museum, Munich

Festival of the Future 2025

Festival of the Future 2025 takes place from July 3 to 6, 2025, covering AI, sustainability, and space, with Björn Ommer as a speaker.


Link to Bayesian Insights on Polarization and Disagreement

Lecture  •  26.06.2025  •  CAS Seestraße 13 80802 Munich

Bayesian Insights on Polarization and Disagreement

As part of the CAS Research Focus, Leah Henderson explores societal polarization using Bayesian models, offering insights on handling disagreements.