A Unifying Review of Linear Gaussian models

摘要

Factor analysis, principal component analysis, mixtures of gaussian clus- ters, vector quantization, Kalman filter models, and hidden Markov mod- els can all be unified as variations of unsupervised learning under a single basic generative model. This is achieved by collecting together disparate observations and derivations made by many previous authors and intro- ducing a new way of linking discrete and continuous stage models using a simple nonlinearity Through the use of other nonlinearities, we show how independent component analysis is also a variation of the same basic generative model.