Guardado en:
Detalles Bibliográficos
Autores principales: Krämer, Fabius, Laux, Tim
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2412.17694
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916634777616384
author Krämer, Fabius
Laux, Tim
author_facet Krämer, Fabius
Laux, Tim
contents We propose and study a novel efficient algorithm for clustering and classification tasks based on the famous MBO scheme. On the one hand, inspired by Jacobs et al. [J. Comp. Phys. 2018], we introduce constraints on the size of clusters leading to a linear integer problem. We prove that the solution to this problem is induced by a novel order statistic. This viewpoint allows us to develop exact and highly efficient algorithms to solve such constrained integer problems. On the other hand, we prove an estimate of the computational complexity of our scheme, which is better than any available provable bounds for the state of the art. This rigorous analysis is based on a variational viewpoint that connects this scheme to volume-preserving mean curvature flow in the big data and small time-step limit.
format Preprint
id arxiv_https___arxiv_org_abs_2412_17694
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle An efficient volume-preserving MBO scheme for data clustering and classification
Krämer, Fabius
Laux, Tim
Analysis of PDEs
Numerical Analysis
Combinatorics
Differential Geometry
68Q25, 90C10, 53E10 (Primary), 58J35, 53Z50, 49Q20, 49Q05
We propose and study a novel efficient algorithm for clustering and classification tasks based on the famous MBO scheme. On the one hand, inspired by Jacobs et al. [J. Comp. Phys. 2018], we introduce constraints on the size of clusters leading to a linear integer problem. We prove that the solution to this problem is induced by a novel order statistic. This viewpoint allows us to develop exact and highly efficient algorithms to solve such constrained integer problems. On the other hand, we prove an estimate of the computational complexity of our scheme, which is better than any available provable bounds for the state of the art. This rigorous analysis is based on a variational viewpoint that connects this scheme to volume-preserving mean curvature flow in the big data and small time-step limit.
title An efficient volume-preserving MBO scheme for data clustering and classification
topic Analysis of PDEs
Numerical Analysis
Combinatorics
Differential Geometry
68Q25, 90C10, 53E10 (Primary), 58J35, 53Z50, 49Q20, 49Q05
url https://arxiv.org/abs/2412.17694