[Identification de paramètre par apprentissage statistique dans un système dynamique modélisant un site de pêche à prix variable]
Dans cette courte note nous étudions les performances de l’apprentissage statistique par réseau de neurones pour l’identification des paramètres d’un modèle de pêche. L’idée est d’observer la pêche pendant quelques jours et d’en déduire les paramètres du modèle et donc la biomasse de poisson sur le long terme.
In this short paper we report on an inverse problem for parameter setting of a model used for the modelling of fishing on the West African coast. We compare the solution of this inverse problem by a Neural Network with the more classical algorithms of optimisation and stochastic control. The Neural Network does much better.
Accepté le :
Publié le :
@article{CRMATH_2020__358_3_245_0, author = {Auger, Pierre and Pironneau, Olivier}, title = {Parameter {Identification} by {Statistical} {Learning} of a {Stochastic} {Dynamical} {System} {Modelling} a {Fishery} with price variation}, journal = {Comptes Rendus. Math\'ematique}, pages = {245--253}, publisher = {Acad\'emie des sciences, Paris}, volume = {358}, number = {3}, year = {2020}, doi = {10.5802/crmath.2}, language = {en}, url = {http://www.numdam.org/articles/10.5802/crmath.2/} }
TY - JOUR AU - Auger, Pierre AU - Pironneau, Olivier TI - Parameter Identification by Statistical Learning of a Stochastic Dynamical System Modelling a Fishery with price variation JO - Comptes Rendus. Mathématique PY - 2020 SP - 245 EP - 253 VL - 358 IS - 3 PB - Académie des sciences, Paris UR - http://www.numdam.org/articles/10.5802/crmath.2/ DO - 10.5802/crmath.2 LA - en ID - CRMATH_2020__358_3_245_0 ER -
%0 Journal Article %A Auger, Pierre %A Pironneau, Olivier %T Parameter Identification by Statistical Learning of a Stochastic Dynamical System Modelling a Fishery with price variation %J Comptes Rendus. Mathématique %D 2020 %P 245-253 %V 358 %N 3 %I Académie des sciences, Paris %U http://www.numdam.org/articles/10.5802/crmath.2/ %R 10.5802/crmath.2 %G en %F CRMATH_2020__358_3_245_0
Auger, Pierre; Pironneau, Olivier. Parameter Identification by Statistical Learning of a Stochastic Dynamical System Modelling a Fishery with price variation. Comptes Rendus. Mathématique, Tome 358 (2020) no. 3, pp. 245-253. doi : 10.5802/crmath.2. http://www.numdam.org/articles/10.5802/crmath.2/
[1] Can overexploited fisheries recover by self-organization? Reallocation of the fishing effort as an emergent form of governance, Marine Policy, Volume 95 (2018), pp. 46-56 | DOI
[2] Deep learning with Python, Manning publications, 2017
[3] Deep Learning, MIT Press, 2016 (http://www.deeplearningbook.org) | Zbl
[4] New development in FreeFem++, J. Numer. Math., Volume 20 (2012) no. 3-4, pp. 251-265 | MR | Zbl
[5] Dynamic programming for mean-field type control, J. Optim. Theory Appl., Volume 169 (2016) no. 3, pp. 902-924 | DOI | MR | Zbl
Cité par Sources :