Field of Study:

Autoencoder is a type of artificial neural network used for learning efficient codings of input data. It's designed to reconstruct its inputs, which forces the model to learn representations of the data in a compressed form.