AutoML in Heavily Constrained Applications

verfasst von

Felix Neutatz, Marius Lindauer, Ziawasch Abedjan

Abstract

Optimizing a machine learning pipeline for a task at hand requires careful configuration of various hyperparameters, typically supported by an AutoML system that optimizes the hyperparameters for the given training dataset. Yet, depending on the AutoML system’s own second-order meta-configuration, the performance of the AutoML process can vary significantly. Current AutoML systems cannot automatically adapt their own configuration to a specific use case. Further, they cannot compile user-defined application constraints on the effectiveness and efficiency of the pipeline and its generation. In this paper, we propose Caml, which uses meta-learning to automatically adapt its own AutoML parameters, such as the search strategy, the validation strategy, and the search space, for a task at hand. The dynamic AutoML strategy of Caml takes user-defined constraints into account and obtains constraint-satisfying pipelines with high predictive performance.

Organisationseinheit(en)

Fachgebiet Maschinelles Lernen
Fachgebiet Datenbanken und Informationssysteme

Externe Organisation(en)

Technische Universität Berlin

Typ

Artikel

Journal

VLDB Journal

Band

33

Seiten

957–979

Anzahl der Seiten

23

ISSN

1066-8888

Publikationsdatum

07.2024

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

ASJC Scopus Sachgebiete

Information systems, Hardware und Architektur

Elektronische Version(en)

https://doi.org/10.48550/arXiv.2306.16913 (Zugang: Offen)
https://doi.org/10.1007/s00778-023-00820-1 (Zugang: Offen)