machine-learning - 在 Keras 中，什么是“密集”层和“辍学”层？

Question

Keras 中密集层和丢失层之间的主要区别是什么

标签： machine-learningkerasartificial-intelligencetransfer-learning

Answer 1

In short, a dropout layer ignores a set of neurons (randomly) as one can see in the picture below. This normally is used to prevent the net from overfitting.

The Dense layer is a normal fully connected layer in a neuronal network.

Resources:

Improving neural networks by preventing co-adaptation of feature detectors

Dropout: A Simple Way to Prevent Neural Networks from Overfitting

A Gentle Introduction to Dropout for Regularizing Deep Neural Networks

Let me know if you need a more precise explanation.