Unified Statistical Theory of Spectral Graph Analysis
This provides a foundational theory for researchers in graph-based machine learning, though it appears incremental as it unifies rather than introduces new methods.
The paper tackles the lack of a unified theoretical framework for spectral graph analysis by recasting it as a nonparametric function estimation problem, resulting in a single formalism and algorithm that accommodates most existing techniques.
The goal of this paper is to show that there exists a simple, yet universal statistical logic of spectral graph analysis by recasting it into a nonparametric function estimation problem. The prescribed viewpoint appears to be good enough to accommodate most of the existing spectral graph techniques as a consequence of just one single formalism and algorithm.