I am currently reading papers about probabilistic PCA and I am wondering why is Gaussian prior (and not some other prior) chosen for the latent variables? Is it just because it's simple or is there another reason?
- Tipping & Bishop, 1999, Probabilistic Principal Component Analysis -- just under eq. (2)
- Tipping & Bishop, 1999, Mixtures of Probabilistic Principal Component Analysers -- eq.(4)