Published February 19, 2024 | Version v1
Publication

Kernel KMeans clustering splits for end-to-end unsupervised decision trees

Others:
Modèles et algorithmes pour l'intelligence artificielle (MAASAI) ; Inria Sophia Antipolis - Méditerranée (CRISAM) ; Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Laboratoire Jean Alexandre Dieudonné (LJAD) ; Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Scalable and Pervasive softwARe and Knowledge Systems (Laboratoire I3S - SPARKS) ; Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S) ; Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis (I3S) ; Université Nice Sophia Antipolis (1965 - 2019) (UNS)-Centre National de la Recherche Scientifique (CNRS)-Université Côte d'Azur (UniCA)-Centre National de la Recherche Scientifique (CNRS)
Centre de recherche du CHU de Québec-Université Laval (CRCHUQ) ; CHU de Québec–Université Laval ; Université Laval [Québec] (ULaval)-Université Laval [Québec] (ULaval)
France Canada Research Fund
ANR-19-P3IA-0002,3IA@cote d'azur,3IA Côte d'Azur(2019)
European Project: H2020-951911,H2020-EU.2.1.1. - INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT),AI4Media(2020)

Description

Trees are convenient models for obtaining explainable predictions on relatively small datasets. Although there are many proposals for the end-to-end construction of such trees in supervised learning, learning a tree end-to-end for clustering without labels remains an open challenge. As most works focus on interpreting with trees the result of another clustering algorithm, we present here a novel end-to-end trained unsupervised binary tree for clustering: Kauri. This method performs a greedy maximisation of the kernel KMeans objective without requiring the definition of centroids. We compare this model on multiple datasets with recent unsupervised trees and show that Kauri performs identically when using a linear kernel. For other kernels, Kauri often outperforms the concatenation of kernel KMeans and a CART decision tree.

Additional details

Created:
March 16, 2024
Modified:
March 16, 2024