Hyper-parameter initialization of classification algorithms using dynamic time warping: A perspective on PCA meta-features

Horváth, Tomáš; Mantovani, Rafael G.; de Carvalho, André C.P.L.F.

Abstract

Meta-learning, a concept from the area of automated machine learning, aims at providing decision support for data scientists by recommending a suitable setting (a machine learning algorithm or its hyper-parameters) to be used for a given dataset. Such a recommendation is based the assumption that an optimal setting for a certain dataset would also be suitable for other, similar datasets. Similarity of datasets is computed from their characteristics, named meta-features, several types of which have been developed thus far. This paper introduces a novel perspective on PCA meta-features which, despite their good descriptive characteristics and easy computation, are rarely used in meta-learning. A novel meta-learning approach utilizing DTW, a well-known similarity measure for time-series, is proposed for computing dataset similarities based on the series of cumulative variances explained by their respective principal components. The results from a large-scale experiment, comparing the proposed approach to multiple baselines on 50 real-world datasets, show the potential of combining PCA and DTW in meta-learning and encourage further investigation in this direction.

• Presentation Conference Type: Conference Paper (published)
• Acceptance Date: Dec 20, 2022
• Online Publication Date: Dec 26, 2022
• Publication Date: 2023-02
• Deposit Date: Mar 27, 2024
• Journal: Applied Soft Computing
• Print ISSN: 1568-4946
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 134
• Article Number: 109969
• DOI: https://doi.org/10.1016/j.asoc.2022.109969
• Keywords: Meta-learning, Meta-features, Principal component analysis, Dynamic time warping, Hyper-parameter initialization
• Public URL: http://researchrepository.napier.ac.uk/Output/3577436