Latent Variable Perspective #

PCA can be interpreted as modelling data using a smaller number of latent variables.

Home | Dimensionality reduction and PCA