« Adadelta » : différence entre les versions
Aucun résumé des modifications |
Aucun résumé des modifications |
||
| Ligne 17 : | Ligne 17 : | ||
== Anglais == | == Anglais == | ||
=== Adadelta=== | |||
Adadelta is a gradient descent based learning algorithm that adapts the learning rate per parameter over time. It was proposed as an improvement over Adagrad, which is more sensitive to hyperparameters and may decrease the learning rate too aggressively. Adadelta It is similar to rmsprop and can be used instead of vanilla SGD. | Adadelta is a gradient descent based learning algorithm that adapts the learning rate per parameter over time. It was proposed as an improvement over Adagrad, which is more sensitive to hyperparameters and may decrease the learning rate too aggressively. Adadelta It is similar to rmsprop and can be used instead of vanilla SGD. | ||
Version du 28 février 2018 à 14:10
Domaine
Vocabulary Apprentissage profond
Définition
Termes privilégiés
Anglais
Adadelta
Adadelta is a gradient descent based learning algorithm that adapts the learning rate per parameter over time. It was proposed as an improvement over Adagrad, which is more sensitive to hyperparameters and may decrease the learning rate too aggressively. Adadelta It is similar to rmsprop and can be used instead of vanilla SGD. • ADADELTA: An Adaptive Learning Rate Method • Stanford CS231n: Optimization Algorithms • An overview of gradient descent optimization algorithms
Contributeurs: Claude Coulombe, Imane Meziani, wiki, Sihem Kouache
