Gated Recurrent Unit (GRU)

De wiki.datafranca.org
Aller à : navigation, rechercher

Domaine

Vocabulary
Claude
Apprentissage profond

Définition



Termes privilégiés

réseau récurrent à portes

Genre: masculin

Choisissez parmi ces termes proposés :
 
un vote
poll-id 47B9DD40F6DE4658BFA4DF28E8632407

Discussion:

Anglais

Gated Recurrent Unit

GRU

The Gated Recurrent Unit is a simplified version of an LSTM unit with fewer parameters. Just like an LSTM cell, it uses a gating mechanism to allow RNNs to efficiently learn long-range dependency by preventing the vanishing gradient problem. The GRU consists of a reset and update gate that determine which part of the old memory to keep vs. update with new values at the current time step. • Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation • Recurrent Neural Network Tutorial, Part 4 – Implementing a GRU/LSTM RNN with Python and Theano