VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models

Understand

Energy-based models (EBMs) have recently been successful in representing complex distributions of small images.

• However, sampling from them requires expensive Markov chain Monte Carlo (MCMC) iterations that mix slowly in high dimensional pixel space.
• Unlike EBMs, variational autoencoders (VAEs) generate samples quickly and are equipped with a latent space that enables fast traversal of the data manifold.
• However, VAEs tend to assign high probability density to regions in data space outside the actual data distribution and often fail at generating sharp images.