Hypergraph $p$-Laplacian equations for data interpolation and semi-supervised learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Kehan, Burger, Martin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910904813092864
author Shi, Kehan
Burger, Martin
author_facet Shi, Kehan
Burger, Martin
contents Hypergraph learning with $p$-Laplacian regularization has attracted a lot of attention due to its flexibility in modeling higher-order relationships in data. This paper focuses on its fast numerical implementation, which is challenging due to the non-differentiability of the objective function and the non-uniqueness of the minimizer. We derive a hypergraph $p$-Laplacian equation from the subdifferential of the $p$-Laplacian regularization. A simplified equation that is mathematically well-posed and computationally efficient is proposed as an alternative. Numerical experiments verify that the simplified $p$-Laplacian equation suppresses spiky solutions in data interpolation and improves classification accuracy in semi-supervised learning. The remarkably low computational cost enables further applications.
format Preprint
id arxiv_https___arxiv_org_abs_2411_12601
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Hypergraph $p$-Laplacian equations for data interpolation and semi-supervised learning
Shi, Kehan
Burger, Martin
Numerical Analysis
Machine Learning
35R02, 65D05
Hypergraph learning with $p$-Laplacian regularization has attracted a lot of attention due to its flexibility in modeling higher-order relationships in data. This paper focuses on its fast numerical implementation, which is challenging due to the non-differentiability of the objective function and the non-uniqueness of the minimizer. We derive a hypergraph $p$-Laplacian equation from the subdifferential of the $p$-Laplacian regularization. A simplified equation that is mathematically well-posed and computationally efficient is proposed as an alternative. Numerical experiments verify that the simplified $p$-Laplacian equation suppresses spiky solutions in data interpolation and improves classification accuracy in semi-supervised learning. The remarkably low computational cost enables further applications.
title Hypergraph $p$-Laplacian equations for data interpolation and semi-supervised learning
topic Numerical Analysis
Machine Learning
35R02, 65D05
url https://arxiv.org/abs/2411.12601