Creating a deep autoencoder using Deep Learning for Java (DL4j)
A deep autoencoder is a deep neural network that is composed of two deep-belief networks that are symmetrical. The networks usually have two separate four or five shallow layers (restricted Boltzmann machines) representing the encoding and decoding half of the net. In this recipe, you will be developing a deep autoencoder consisting of one input layer, four decoding layers, four encoding layers, and one output layer. In doing so, we will be using a very popular dataset named MNIST.