Position: Why We Must Rethink Empirical Research in Machine Learning
Machine Learning
2024-05-28 v2 Machine Learning
Abstract
We warn against a common but incomplete understanding of empirical research in machine learning that leads to non-replicable results, makes findings unreliable, and threatens to undermine progress in the field. To overcome this alarming situation, we call for more awareness of the plurality of ways of gaining knowledge experimentally but also of some epistemic limitations. In particular, we argue most current empirical machine learning research is fashioned as confirmatory research while it should rather be considered exploratory.
Cite
@article{arxiv.2405.02200,
title = {Position: Why We Must Rethink Empirical Research in Machine Learning},
author = {Moritz Herrmann and F. Julian D. Lange and Katharina Eggensperger and Giuseppe Casalicchio and Marcel Wever and Matthias Feurer and David Rügamer and Eyke Hüllermeier and Anne-Laure Boulesteix and Bernd Bischl},
journal= {arXiv preprint arXiv:2405.02200},
year = {2024}
}
Comments
20 pages, accepted for publication at ICML 2024, camera-ready version