Titre :

Comparing Parallel Surrogate-based and Surrogate-free Multi-Objective Optimization of COVID-19 vaccines allocation

Auteur(s) :

Briffoteaux, Guillaume [Auteur]
Université de Mons / University of Mons [UMONS]
Optimisation de grande taille et calcul large échelle [BONUS]
Ragonnet, Romain [Auteur]
Monash University [Melbourne]
Tomenko, Pierre [Auteur]
Université de Mons / University of Mons [UMONS]
Mezmaz, Mohand [Auteur]
Université de Mons / University of Mons [UMONS]
Melab, Nouredine [Auteur]
Optimisation de grande taille et calcul large échelle [BONUS]
Tuyttens, Daniel [Auteur]
Université de Mons / University of Mons [UMONS]

Titre de la manifestation scientifique :

OLA 2022 - International Conference on Optimization and Learning

Ville :

Syracuse

Pays :

Italie

Date de début de la manifestation scientifique :

2022-07-18

Discipline(s) HAL :

Mathématiques [math]/Optimisation et contrôle [math.OC]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Modélisation et simulation

Résumé en anglais : [en]

The simulation-based and computationally expensive problem tackled in this paper addresses COVID-19 vaccines allocation in Malaysia. The multi-objective formulation considers simultaneously the total number of deaths, peak ...
Lire la suite >The simulation-based and computationally expensive problem tackled in this paper addresses COVID-19 vaccines allocation in Malaysia. The multi-objective formulation considers simultaneously the total number of deaths, peak hospital occupancy and relaxation of mobility restrictions. Evolutionary algorithms have proven their capability to handle multi-to-many objectives but require a high number of computationally expensive simulations. The available techniques to raise the challenge rely on the joint use of surrogate-assisted optimization and parallel computing to deal with computational expensiveness. On the one hand, the simulation software is imitated by a cheap-to-evaluate surrogate model. On the other hand, multiple candidates are simultaneously assessed via multiple processing cores. In this study, we compare the performance of recently proposed surrogate-free and surrogate-based parallel multi-objective algorithms through the application to the COVID-19 vaccine distribution problem.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL