A Convex Formulation for Learning Scale-Free Networks via Submodular Relaxation
This work addresses network structure determination in statistics and machine learning, specifically for scale-free graphs, offering a method that is incremental in its approach.
The paper tackles the problem of reconstructing scale-free network structures from data by formulating structured sparsity priors using submodular functions and their Lovász extension for convex relaxation. It results in improved accuracy for synthetic data and encourages scale-free reconstructions on a bioinformatics dataset.
A key problem in statistics and machine learning is the determination of network structure from data. We consider the case where the structure of the graph to be reconstructed is known to be scale-free. We show that in such cases it is natural to formulate structured sparsity inducing priors using submodular functions, and we use their Lovász extension to obtain a convex relaxation. For tractable classes such as Gaussian graphical models, this leads to a convex optimization problem that can be efficiently solved. We show that our method results in an improvement in the accuracy of reconstructed networks for synthetic data. We also show how our prior encourages scale-free reconstructions on a bioinfomatics dataset.