Landmark Alternating Diffusion
This work addresses computational limitations in sensor fusion for applications like sleep stage annotation, but it is incremental as it builds on existing diffusion methods.
The authors tackled the computational burden of Alternating Diffusion (AD) by proposing Landmark AD (LAD), which offers superior computational efficiency while capturing AD's essence, as demonstrated in automatic sleep stage annotation with two EEG channels.
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 manifold setup and apply it to the automatic sleep stage annotation problem with two electroencephalogram channels to demonstrate its application.