Date/heure
Date(s) - 22 juin 2018

Catégories Pas de Catégories

$I will present an extension of the cut-pursuit algorithm, introduced by Landrieu and Obozinski (2017), to the _graph total-variation_ regularization of functions with a separable nondifferentiable part.We propose a modified algorithmic scheme as well as adapted proofs of convergence. We also present a heuristic approach for handling the cases in which the values associated to each vertex of the graph are multidimensional.The performance of our algorithm, which we demonstrate on difficult, ill-conditioned large-scale inverse and learning problems, is such that it may in practice extend the scope of application of the total-variation regularization. -This is a joint work with Loïc Landrieu (IGN, LaSTIG MATIS). -Landrieu, L. and Obozinski, G. “Cut pursuit : Fast algorithms to learn piecewiseconstant functions on general weighted graphs”. SIAM Journal on ImagingSciences, 10(4):1724–1766, 2017. https://www.researchgate.net/scientific-contributions/2046330849_Hugo_Raguet http://1a7r0ch3.github.io Hugo RAGUET [$