TRR 318 - Project C03: Interpretable machine learning: Explaining change

Overview

Project C03 centers on how to account for dynamic properties of an explanandum. Its target is to deliver a model that will explain a drift of data that a machine learning model undergoes over time as a result of nonstationary distributions and evolving training data. The project will go beyond a mere transfer of established methodology for interpretable ML and develop a new methodology for explainable ML in the extremely relevant area of online learning. Questions to be addressed are: How to keep the frequency of model change at a reasonably low level while guaranteeing a sufficient quality and accuracy, and how to justify a model change and explain it to the user.

Key Facts

Project type:
Forschung
Project duration:
07/2021 - 12/2025

More Information

Principal Investigators

contact-box image

Externe Projektleitung

contact-box image

Barbara Hammer

Universität Bielefeld

Cooperating Institutions

Ludwig-Maximilian-Universität München

Cooperating Institution

Universität Bielefeld

Cooperating Institution

Publications

Exact Computation of Any-Order Shapley Interactions for Graph Neural Networks
M. Muschalik, F. Fumagalli, P. Frazzetto, J. Strotherm, L. Hermes, A. Sperduti, E. Hüllermeier, B. Hammer, in: The Thirteenth International Conference on Learning Representations (ICLR), 2025.
Explaining Outliers using Isolation Forest and Shapley Interactions
R. Visser, F. Fumagalli, E. Hüllermeier, B. Hammer, in: Proceedings of the European Symposium on Artificial Neural Networks (ESANN), 2025.
Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theory
F. Fumagalli, M. Muschalik, E. Hüllermeier, B. Hammer, J. Herbinger, in: Proceedings of The 28th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR, 2025, pp. 5140–5148.
Adaptive Prompting: Ad-hoc Prompt Composition for Social Bias Detection
M. Spliethöver, T. Knebler, F. Fumagalli, M. Muschalik, B. Hammer, E. Hüllermeier, H. Wachsmuth, in: L. Chiruzzo, A. Ritter, L. Wang (Eds.), Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Association for Computational Linguistics, Albuquerque, New Mexico, 2025, pp. 2421–2449.
No learning rates needed: Introducing SALSA - Stable Armijo Line Search Adaptation
P. Kenneweg, T. Kenneweg, F. Fumagalli, B. Hammer, in: 2024 International Joint Conference on Neural Networks (IJCNN), 2024, pp. 1–8.
Show all publications