Title: Characterization and Inference of Graph Diffusion Processes from Observations of Stationary Signals

Abstract: Many tools from the field of graph signal processing exploit knowledge of the underlying graph's structure (e.g., as encoded in the Laplacian matrix) to process signals on the graph. Therefore, in the case when no graph is available, graph signal processing tools cannot be used anymore. Researchers have proposed approaches to infer a graph topology from observations of signals on its nodes. Since the problem is ill-posed, these approaches make assumptions on the graph to be recovered, such as smoothness of the signals on the graph, or sparsity priors. In this paper, we propose a characterization of the space of valid graphs, in the sense that they can explain stationary signals. To simplify the exposition in this paper, we focus here on the case where signals were i.i.d at some point back in time and were observed after diffusion on a graph. We show that the set of graphs verifying this assumption has a strong connection with the eigenvectors of the covariance matrix, and forms a convex space. Along with a theoretical study in which these eigenvectors are assumed to be known, we consider the practical case when the observations are noisy, and experimentally observe how fast the space of valid graphs converges to the space obtained when the exact eigenvectors are known as the number of observations grows. To illustrate how this characterization can be used for graph recovery, two methods for selecting a particular point in this space under chosen criteria are presented, namely graph simplicity and sparsity. Finally, we evaluate how state-of-the-art methods relate to this framework through experiments on a real-life dataset of temperatures in Brittany, France.