Autoencoders as Neural Network Version of PCA
Explore how autoencoders serve as neural network equivalents of principal component analysis to perform dimension reduction and feature extraction. Understand their encoding and decoding processes, their relation to PCA, and the use of tied weights. This lesson equips you with knowledge about autoencoders' role in anomaly detection and unsupervised learning applications.
Background
An autoencoder is a reconstruction model. It attempts to reconstruct its inputs from itself as depicted below:
An autoencoder is made of two modules: encoder and decoder.
As their names indicate, an encoder encodes input ...