A generic coordinate descent solver for non-smooth convex optimisation

Autor: Olivier Fercoq

Publikováno v: Optimization Methods and Software
Optimization Methods and Software, Taylor & Francis, 2019, pp.1-21. ⟨10.1080/10556788.2019.1658758⟩

International audience; We present a generic coordinate descent solver for the minimization of a nonsmooth convex objective with structure. The method can deal in particular with problems with linear constraints. The implementation makes use of effic

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3736bb66014c9efb9a39a29e68fdddda
https://doi.org/10.1080/10556788.2019.1658758

Feature selection algorithms for flexible analog-to-feature converter

Autor: Antoine Back, Patricia Desgreys, Paul Chollet, Olivier Fercoq

Publikováno v: 2020 18th IEEE International New Circuits and Systems Conference (NEWCAS)
2020 18th IEEE International New Circuits and Systems Conference (NEWCAS), Jun 2020, Montréal, Canada. pp.186-189, ⟨10.1109/NEWCAS49341.2020.9159817⟩
NEWCAS

International audience; One of the main challenges in the field of wireless sensors is to increase their battery life. Analog-to-feature (A2F) conversion is an acquisition method thought for IoT devices, that perform classification tasks at sub-Nyqui

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6ec324e9885ba3645955b5c8a94d5dc7
https://hal.archives-ouvertes.fr/hal-02927208

Benchmarking GNN-CMA-ES on the BBOB noiseless testbed

Autor: Olivier Fercoq, Louis Faury, Clément Calauzènes

Publikováno v: the Genetic and Evolutionary Computation Conference Companion
the Genetic and Evolutionary Computation Conference Companion, Jul 2019, Prague, Czech Republic. pp.1928-1936
GECCO (Companion)

International audience; Popular machine learning estimators involve regularization parameters that can be challenging to tune, and standard strategies rely on grid search for this task. In this paper, we revisit the techniques of approximating the re

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d5a877a1baf40fe865efb9fd3e785c97
https://hal.archives-ouvertes.fr/hal-02269139

A Coordinate Descent Primal-Dual Algorithm with Large Step Size and Possibly Non Separable Functions

Autor: Pascal Bianchi, Olivier Fercoq

Publikováno v: SIAM Journal on Optimization
SIAM Journal on Optimization, Society for Industrial and Applied Mathematics, 2019, 29 (1), pp.100-134

This paper introduces a coordinate descent version of the V\~u-Condat algorithm. By coordinate descent, we mean that only a subset of the coordinates of the primal and dual iterates is updated at each iteration, the other coordinates being maintained

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b5a944b87e6c977b50a4b16275c0a7f8
https://hal.archives-ouvertes.fr/hal-01497104

Using big steps in coordinate descent primal-dual algorithms

Autor: Pascal Bianchi, Olivier Fercoq

Publikováno v: IEEE 55th Conference on Decision and Control (CDC), 2016
IEEE 55th Conference on Decision and Control (CDC)
IEEE 55th Conference on Decision and Control (CDC), Dec 2016, Las Vegas, NV, United States. pp.1895-1899, ⟨10.1109/CDC.2016.7798541⟩
CDC

International audience; The Vu-Condat algorithm is a standard method for finding a saddle point of a Lagrangian involving a differentiable function. Recent works have tried to adapt the idea of random coordinate descent to this algorithm, with the ai

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::eab72372a3fb88ccf99f722d792db71d
https://hal.archives-ouvertes.fr/hal-01497087/file/cdc.pdf