par Loris, Ignace 
Référence 9th International Conference ”Inverse Problems: Modeling and Simulation” (21--25/05/2018: Malta)
Publication Non publié, 2018-05-22
Référence 9th International Conference ”Inverse Problems: Modeling and Simulation” (21--25/05/2018: Malta)
Publication Non publié, 2018-05-22
Communication à un colloque
| Résumé : | We show convergence of a number of iterative optimization algorithms consisting of nested primal-dual proximal gradient iterations. The algorithms can be used for the numerical solution of convex optimization problems defined by the sum of a differentiable term and two possibly non-differentiable terms. One of the latter terms should take the form of the composition of a linear map and a proximable function, while the differentiable term needs an accessible gradient. The algorithms reduce to the usual proximal gradient algorithm in certain special cases and generalize other existing algorithms. In addition, under some conditions of strong convexity, we show a linear rate of convergence. Numerical experiments illustrate their applicability to large-scale optimization problems in mathematical imaging, in particular to GPS tomography. |
