University Sétif 1 FERHAT ABBAS Faculty of Sciences
Détail de l'auteur
Auteur leila Djidel |
Documents disponibles écrits par cet auteur
Ajouter le résultat dans votre panier Affiner la recherche
Titre : Tuning deep learning parameters using bio-inspired algorithm Type de document : texte imprimé Auteurs : Aya Lakhdari, Auteur ; leila Djidel, Auteur ; Semcheddine,Moussa, Directeur de thèse Année de publication : 2022 Importance : 1 vol (67 f .) Format : 29cm Langues : Français (fre) Catégories : Thèses & Mémoires:Informatique Mots-clés : Informatique
InternetIndex. décimale : 004 Informatique Résumé :
There are several domains in which deep learning techniques are successfully applied.
The great success of deep learning depends on the great performance of models. Learning
rate, choice of activation function, network architectures..., all of these are deep learning
hyper-parameters that had an important role in improving the performance of models,
but there are no rules for choosing the right value of each hyper-parameter. In this
thesis, we will use some bio-inspired algorithms like genetic algorithm, particle swarm
optimization algorithm, black widow optimization algorithm, and grey wolf optimization
algorithm to optimize the CNN and LSTM models hyper-parameters. We have used two
data-set, the first is the MNIST data-set which concerned images and the second is the
20 newsgroups text data-set. And we have illustrated the results of the tests affected by
the bio-inspired algorithms and both random and grid searches. The results show that
bio-inspired algorithms give better accuracy results so it can be the way to tune deep
learning hyper-parameters.Côte titre : MAI/0675 En ligne : https://drive.google.com/file/d/1Jkj-sFnSsaxyLtTPAkh1Gz2q1HRUgbHK/view?usp=share [...] Format de la ressource électronique : Tuning deep learning parameters using bio-inspired algorithm [texte imprimé] / Aya Lakhdari, Auteur ; leila Djidel, Auteur ; Semcheddine,Moussa, Directeur de thèse . - 2022 . - 1 vol (67 f .) ; 29cm.
Langues : Français (fre)
Catégories : Thèses & Mémoires:Informatique Mots-clés : Informatique
InternetIndex. décimale : 004 Informatique Résumé :
There are several domains in which deep learning techniques are successfully applied.
The great success of deep learning depends on the great performance of models. Learning
rate, choice of activation function, network architectures..., all of these are deep learning
hyper-parameters that had an important role in improving the performance of models,
but there are no rules for choosing the right value of each hyper-parameter. In this
thesis, we will use some bio-inspired algorithms like genetic algorithm, particle swarm
optimization algorithm, black widow optimization algorithm, and grey wolf optimization
algorithm to optimize the CNN and LSTM models hyper-parameters. We have used two
data-set, the first is the MNIST data-set which concerned images and the second is the
20 newsgroups text data-set. And we have illustrated the results of the tests affected by
the bio-inspired algorithms and both random and grid searches. The results show that
bio-inspired algorithms give better accuracy results so it can be the way to tune deep
learning hyper-parameters.Côte titre : MAI/0675 En ligne : https://drive.google.com/file/d/1Jkj-sFnSsaxyLtTPAkh1Gz2q1HRUgbHK/view?usp=share [...] Format de la ressource électronique : Exemplaires (1)
Code-barres Cote Support Localisation Section Disponibilité MAI/0675 MAI/0675 Mémoire Bibliothéque des sciences Anglais Disponible
Disponible