Autoencoding is a data compression algorithm where the compression and decompression functions are
- data-specific
- lossy
- learned automatically from examples rather than engineered by a human.
In almost all contexts where the term Autoencoder is used, the compression and decompression functions are implemented with neural networks.
The simples possible Autoencoder in which the encoder and decoder consists of a single fully-connected layer each.
A Deep Autoencoder in which both the encoder and decoder consists of multiple fully-connected layer each.
A Deep Autoencoder in which both the encoder and decoder consists of multiple fully-convolutional layer each.
A fully convolutional Autoencoder for Denoising images with noisy backgrounds.