Accelerated Graph Learning From Smooth Signals

We consider network topology identification subject to a signal smoothness prior on the nodal observations. A fast dual-based proximal gradient algorithm is developed efficiently tackle strongly convex, smoothness-regularized inverse problem known yield high-quality graph solutions. Unlike existing solvers, novel iterations come with global convergence rate guarantees and do not require additional step-size tuning. Reproducible simulated tests demonstrate effectiveness of proposed method in accurately recovering random real-world graphs, markedly faster than state-of-the-art alternatives without incurring an extra computational burden.