Passer au contenu principal

Vocabulary of natural language processing (POC)

Choisissez le vocabulaire dans lequel chercher

Concept information

Terme préférentiel

backpropagation  

Définition

  • A gradient estimation technique used in supervised machine learning which consists in updating the weights of the neural network by iteratively propagating the error backward from the output layer to the input layer and then minimize the difference between the predicted and actual output. The backpropagation algorithm is used in image, speech, and pattern recognition as well as classification.

Concept générique

Synonyme(s)

  • backprop
  • backward propagation of errors
  • BP
  • error back propagation
  • error feedback propagation
  • feedback propagation

Note

  • The term back-propagation is often misunderstood as meaning the whole learning algorithm for multilayer neural networks. Backpropagation refers only to the method for computing the gradient, while other algorithms, such as stochastic gradient descent, is used to perform learning using this gradient. (Goodfellow et al., 6.5 Back-Propagation and Other Differentiation Algorithms, Deep Learning, 2016)

Traductions

  • français

  • rétropropagation d'erreur
  • rétropropagation d'erreurs
  • rétropropagation de l'erreur
  • rétropropagation des erreurs

URI

http://data.loterre.fr/ark:/67375/8LP-M7H9HK89-Z

Télécharger ce concept :

RDF/XML TURTLE JSON-LD Dernière modification le 02/05/2024