VAE: Latent distribution. Posterior collapse, Multiple latents
I have two questions after exploring VAE for a while. In the standard VAE setup, we assume 1 latent variable of shape (BHWD): mu and var, and the prior N(0, I).
difference between autoencoder and variational autoencoder
I was in the process of learning about variation autoencoders (VAE) but couldn’t quite grasp the concept. I would appreciate it if you could help me understand it better.