Q(D)O-ES: Population-based Quality (Diversity) Optimisation for Post Hoc Ensemble Selection in AutoML

Purucker, Lennart Oswald; Schneider, Lennart ; Anastacio, Marie; Beel, Joeran; Bischl, Bernd und Hoos, Holger : Q(D)O-ES: Population-based Quality (Diversity) Optimisation for Post Hoc Ensemble Selection in AutoML. 2nd International Conference on Automated Machine Learning (AutoML), Potsdam, Germany, 12 - 15 November 2023. Faust, Aleksandra; Garnett, Roman; White, Colin; Hutter, Frank und Gardner, Jacob R. (Hrsg.): In: Proceedings of the Second International Conference on Automated Machine Learning, Proceedings of Machine Learning Research Bd. 224, 10 MLResearchPress.

Abstract

Automated machine learning (AutoML) systems commonly ensemble models post hoc to improve predictive performance, typically via greedy ensemble selection (GES). However, we believe that GES may not always be optimal, as it performs a simple deterministic greedy search. In this work, we introduce two novel population-based ensemble selection methods, QO-ES and QDO-ES, and compare them to GES. While QO-ES optimises solely for predictive performance, QDO-ES also considers the diversity of ensembles within the population, maintaining a diverse set of well-performing ensembles during optimisation based on ideas of quality diversity optimisation. The methods are evaluated using 71 classification datasets from the AutoML benchmark, demonstrating that QO-ES and QDO-ES often outrank GES, albeit only statistically significant on validation data. Our results further suggest that diversity can be beneficial for post hoc ensembling but also increases the risk of overfitting.

• Dokumententyp: Konferenzbeitrag (Paper)
• Fakultät: Mathematik, Informatik und Statistik > Statistik > Lehrstühle/Arbeitsgruppen > Computationale Statistik
• Themengebiete: 000 Informatik, Informationswissenschaft, allgemeine Werke > 004 Informatik ; 500 Naturwissenschaften und Mathematik > 510 Mathematik
• ISSN: 2640-3498
• Sprache: Englisch