AI Glossary/Dropout
Deep Learning

Dropout

A regularization technique that randomly drops neurons during training to prevent overfitting.

In-depth explanation

During each training step, dropout randomly sets a fraction of neuron outputs to zero. This prevents neurons from co-adapting too much and forces the network to learn more robust features. At inference time, all neurons are used but outputs are scaled. Dropout rate typically ranges from 0.1 to 0.5.

Examples

Dropout rate of 0.5 in dense layers
Spatial dropout in CNNs

Related terms

Master Dropout.

Learn how to apply this concept with hands-on projects in our comprehensive AI programs.