Landmark Alternating Diffusion

arXiv:2404.19649v1 Announce Type: new
Abstract: Alternating Diffusion (AD) is a commonly applied diffusion-based sensor fusion algorithm. While it has been successfully applied to various problems, its computational burden remains a limitation. Inspired by the landmark diffusion idea 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 …

abstract algorithm arxiv computational cs.lg diffusion embedding fusion landmark math.st physics.data-an robust scalable sensor stat.ml stat.th type variation via while