Hyperbolic Random Forests

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Doorenbos, Lars, Márquez-Neila, Pablo, Sznitman, Raphael, Mettes, Pascal
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916298262315008
author Doorenbos, Lars
Márquez-Neila, Pablo
Sznitman, Raphael
Mettes, Pascal
author_facet Doorenbos, Lars
Márquez-Neila, Pablo
Sznitman, Raphael
Mettes, Pascal
contents Hyperbolic space is becoming a popular choice for representing data due to the hierarchical structure - whether implicit or explicit - of many real-world datasets. Along with it comes a need for algorithms capable of solving fundamental tasks, such as classification, in hyperbolic space. Recently, multiple papers have investigated hyperbolic alternatives to hyperplane-based classifiers, such as logistic regression and SVMs. While effective, these approaches struggle with more complex hierarchical data. We, therefore, propose to generalize the well-known random forests to hyperbolic space. We do this by redefining the notion of a split using horospheres. Since finding the globally optimal split is computationally intractable, we find candidate horospheres through a large-margin classifier. To make hyperbolic random forests work on multi-class data and imbalanced experiments, we furthermore outline a new method for combining classes based on their lowest common ancestor and a class-balanced version of the large-margin loss. Experiments on standard and new benchmarks show that our approach outperforms both conventional random forest algorithms and recent hyperbolic classifiers.
format Preprint
id arxiv_https___arxiv_org_abs_2308_13279
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Hyperbolic Random Forests
Doorenbos, Lars
Márquez-Neila, Pablo
Sznitman, Raphael
Mettes, Pascal
Machine Learning
Artificial Intelligence
Hyperbolic space is becoming a popular choice for representing data due to the hierarchical structure - whether implicit or explicit - of many real-world datasets. Along with it comes a need for algorithms capable of solving fundamental tasks, such as classification, in hyperbolic space. Recently, multiple papers have investigated hyperbolic alternatives to hyperplane-based classifiers, such as logistic regression and SVMs. While effective, these approaches struggle with more complex hierarchical data. We, therefore, propose to generalize the well-known random forests to hyperbolic space. We do this by redefining the notion of a split using horospheres. Since finding the globally optimal split is computationally intractable, we find candidate horospheres through a large-margin classifier. To make hyperbolic random forests work on multi-class data and imbalanced experiments, we furthermore outline a new method for combining classes based on their lowest common ancestor and a class-balanced version of the large-margin loss. Experiments on standard and new benchmarks show that our approach outperforms both conventional random forest algorithms and recent hyperbolic classifiers.
title Hyperbolic Random Forests
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2308.13279