Landmark Alternating Diffusion
Journal
SIAM Journal on Mathematics of Data Science
Journal Volume
7
Journal Issue
2
Start Page
621
End Page
642
ISSN
2577-0187
Date Issued
2025-05-07
Author(s)
Abstract
Alternating Diffusion (AD) is a commonly used diffusion-based sensor fusion algorithm. While it has been successfully applied to various problems, its computational burden remains a limitation. Inspired by the use of landmarks considered in the Robust and Scalable Embedding via Landmark Diffusion (ROSELAND), we propose a variation of AD, called Landmark AD (LAD), which captures the essence of AD while offering superior computational efficiency. We provide a series of theoretical analyses of LAD under the manifold setup and apply it to the automatic sleep stage annotation problem with two electroencephalogram channels to demonstrate its application.
Publisher
Society for Industrial & Applied Mathematics (SIAM)
Type
journal article