The article introduces CURSOR, a novel framework for hypergraph matching that leverages CUR tensor decomposition. By utilizing a cascaded second and third-order approach, CURSOR significantly reduces time complexity and tensor density in large-scale graph matching. The method integrates seamlessly into existing algorithms, improving performance while lowering computational costs. Experimental results demonstrate the superiority of CURSOR over traditional methods on synthetic datasets and benchmark sets.
To Another Language
from source content
arxiv.org
Viktige innsikter hentet fra
by Qixuan Zheng... klokken arxiv.org 03-15-2024
https://arxiv.org/pdf/2402.16594.pdfDypere Spørsmål