Afficher la notice abrégée

EpidemiOptim: a Toolbox for the Optimization of Control Policies in Epidemiological Models

dc.rights.licenseopenen_US
dc.contributor.authorCOLAS, Cedric
hal.structure.identifierStatistics In System biology and Translational Medicine [SISTM]
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorHEJBLUM, Boris
ORCID: 0000-0003-0646-452X
IDREF: 189970316
hal.structure.identifierGroupe de Recherche en Economie Théorique et Appliquée [GREThA]
dc.contributor.authorROUILLON, Sebastien
IDREF: 149491913
hal.structure.identifierStatistics In System biology and Translational Medicine [SISTM]
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorTHIEBAUT, Rodolphe
dc.contributor.authorOUDEYER, Pierre-Yves
dc.contributor.authorMOULIN-FRIER, Clement
hal.structure.identifierStatistics In System biology and Translational Medicine [SISTM]
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorPRAGUE, Melanie
dc.date.accessioned2021-05-07T12:28:45Z
dc.date.available2021-05-07T12:28:45Z
dc.date.created2020
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/27206
dc.description.abstractEnModelling the dynamics of epidemics helps proposing control strategies based on phar-maceutical and non-pharmaceutical interventions (contact limitation, lock down, vaccina-tion, etc). Hand-designing such strategies is not trivial because of the number of pos-sible interventions and the difficulty to predict long-term effects. This task can be castas an optimization problem where state-of-the-art machine learning algorithms such asdeep reinforcement learning, might bring significant value. However, the specificity ofeach domain – epidemic modelling or solving optimization problem – requires strong col-laborations between researchers from different fields of expertise. This is why we intro-duce EpidemiOptim, a Python toolbox that facilitates collaborations between researchersin epidemiology and optimization. EpidemiOptim turns epidemiological models and costfunctions into optimization problems via a standard interface commonly used by optimiza-tion practitioners (OpenAI Gym). Reinforcement learning algorithms based on Q-Learningwith deep neural networks (dqn) and evolutionary algorithms (nsga-ii) are already im-plemented. We illustrate the use of EpidemiOptim to find optimal policies for dynamicalon-off lock-down control under the optimization of death toll and economic recess using aSusceptible-Exposed-Infectious-Removed (seir) model for COVID-19. Using EpidemiOp-tim and its interactive visualization platform in Jupyter notebooks, epidemiologists, op-timization practitioners and others (e.g. economists) can easily compare epidemiologicalmodels, costs functions and optimization algorithms to address important choices to bemade by health decision-makers. Trained models can be explored by experts and non-experts via a web interface.
dc.language.isoENen_US
dc.title.enEpidemiOptim: a Toolbox for the Optimization of Control Policies in Epidemiological Models
dc.typeDocument de travail - Pré-publicationen_US
dc.subject.halInformatique [cs]/Apprentissage [cs.LG]en_US
dc.subject.halInformatique [cs]/Intelligence artificielle [cs.AI]en_US
dc.subject.halSciences du Vivant [q-bio]/Bio-Informatique, Biologie Systémique [q-bio.QM]en_US
bordeaux.hal.laboratoriesBordeaux Population Health Research Center (BPH) - UMR 1219en_US
bordeaux.institutionUniversité de Bordeauxen_US
bordeaux.institutionINSERMen_US
bordeaux.teamSISTMen_US
bordeaux.teamSISTM_BPH
bordeaux.import.sourcehal
hal.identifierhal-03099898
hal.version1
hal.exportfalse
workflow.import.sourcehal
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.au=COLAS,%20Cedric&HEJBLUM,%20Boris&ROUILLON,%20Sebastien&THIEBAUT,%20Rodolphe&OUDEYER,%20Pierre-Yves&rft.genre=preprint


Fichier(s) constituant ce document

FichiersTailleFormatVue

Il n'y a pas de fichiers associés à ce document.

Ce document figure dans la(les) collection(s) suivante(s)

Afficher la notice abrégée