Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Computation - Neural C omputation is the subject of a report. According to news reporting out of Naples , Italy, by NewsRx editors, research stated, “We propose and analyze a continuou s-time firing-rate neural network, the positive firing-rate competitive network (PFCN), to tackle sparse reconstruction problems with nonnegativity constraints. These problems, which involve approximating a given input stimulus from a dicti onary using a set of sparse (active) neurons, play a key role in a wide range of domains, including, for example, neuroscience, signal processing, and machine l earning.” Our news journalists obtained a quote from the research, “First, by leveraging t he theory of proximal operators, we relate the equilibria of a family of continu ous-time firing-rate neural networks to the optimal solutions of sparse reconstr uction problems. Then we prove that the PFCN is a positive system and give rigor ous conditions for the convergence to the equilibrium. Specifically, we show tha t the convergence depends only on a property of the dictionary and is linear-exp onential in the sense that initially, the convergence rate is at worst linear an d then, after a transient, becomes exponential. We also prove a number of techni cal results to assess the contractivity properties of the neural dynamics of int erest. Our analysis leverages contraction theory to characterize the behavior of a family of firing-rate competitive networks for sparse reconstruction with and without nonnegativity constraints.”
NETL
NSTL