Deep Learning Specialization - Coursera

main

main

Introduction
Neural Networks and Deep Learning
Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization
Structuring Machine Learning Projects
Convolutional Neural Networks
Sequence Models

Powered by GitBook

On this page

Was this helpful?

Sequence Models

Recurrent Neural Networks

Gated Recurrent Units (GRUs)

PreviousVanishing Gradient Problem in RNNs NextLong Short-Term Memory Network (LSTM)

Last updated 4 years ago

Was this helpful?

GRUs help solve the vanishing gradient problem in RNNs. They allow for long-range dependencies.

The GRU unit is shown below:

The following are the equations associated with a single GRU unit.

r stands for the reset gate and u stands for update gate.