February 2019
Beginner to intermediate
308 pages
7h 42m
English
In Chapter 5, Removing Noise from Images Using Autoencoders, we looked at autoencoders, a special class of neural networks that learns a latent representation of the input. Autoencoders have an encoder component that compresses the input into a latent representation, and a decoder component that reconstructs the input using the latent representation.
In an autoencoder, the size of the hidden layer that's used for the latent representation is an important hyperparameter that needs to be tuned carefully. The size of the latent representation should be sufficiently small enough to represent a compressed representation of the input features, and also sufficiently large enough for the decoder to reconstruct ...