Graph topology inference based on sparsifying transform learning

TL;DR

This paper introduces a novel method for inferring graph topology by learning a sparsifying transform, enabling the construction of modular graphs with eigenvectors aligned to the data, demonstrated on synthetic and brain data.

Contribution

It proposes a two-step optimization approach combining transform learning and convex optimization to infer graph structures from data, advancing graph signal processing techniques.

Findings

Effective on synthetic data

Successful application to brain data

Outperforms existing methods in topology inference

Abstract

Graph-based representations play a key role in machine learning. The fundamental step in these representations is the association of a graph structure to a dataset. In this paper, we propose a method that aims at finding a block sparse representation of the graph signal leading to a modular graph whose Laplacian matrix admits the found dictionary as its eigenvectors. The role of sparsity here is to induce a band-limited representation or, equivalently, a modular structure of the graph. The proposed strategy is composed of two optimization steps: i) learning an orthonormal sparsifying transform from the data; ii) recovering the Laplacian, and then topology, from the transform. The first step is achieved through an iterative algorithm whose alternating intermediate solutions are expressed in closed form. The second step recovers the Laplacian matrix from the sparsifying transform through a convex optimization method. Numerical results corroborate the effectiveness of the proposed methods over both synthetic data and real brain data, used for inferring the brain functionality network through experiments conducted over patients affected by epilepsy.