A convex formulation for learning scale-free networks via submodular relaxation

Defazio, AaronCaetano, Tiberio2015-12-13December 39781627480031http://hdl.handle.net/1885/82731A 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 formulatAuthor/s retain copyrightKeywords: Convex optimization problems; Convex relaxation; Gaussian graphical models; Network structures; Scale free networks; Structured sparsities; Submodular functions; Tractable class; Convex optimization; Relaxation processesA convex formulation for learning scale-free networks via submodular relaxation20122016-02-24