An Autoencoder is a type of artificial neural network used to learn efficient data codings in an unsupervised manner.
At its core, an autoencoder is designed to compress input data into a lower-dimensional code and then reconstruct the output from this representation. The goal is for the output to be as close to the original input as possible.
How an Autoencoder Works
An autoencoder consists of three main components:
- The Encoder: This part of the network compresses the input into a “latent-space representation.” It progressively reduces the dimensions of the data.
- The Code (Bottleneck): This is the hidden layer that contains the compressed representation of the input. This is the “knowledge” the network has extracted.
- The Decoder: This part of the network mirrors the encoder. It takes the compressed code and attempts to reconstruct the original data.
The network is trained by minimizing the reconstruction error, which is the difference between the original input (X) and the reconstructed output (hat X):