Improving Sampling from Generative Autoencoders with Markov Chains
This addresses a specific bottleneck in generative modeling for researchers, offering an incremental improvement in sampling techniques.
The paper tackles the problem of sampling from generative autoencoders when the learned latent distribution deviates from the prior, proposing a Markov chain Monte Carlo (MCMC) method that iteratively decodes and encodes to sample from this distribution, which improves sample quality and reveals differences between models trained with or without denoising.
We focus on generative autoencoders, such as variational or adversarial autoencoders, which jointly learn a generative model alongside an inference model. Generative autoencoders are those which are trained to softly enforce a prior on the latent distribution learned by the inference model. We call the distribution to which the inference model maps observed samples, the learned latent distribution, which may not be consistent with the prior. We formulate a Markov chain Monte Carlo (MCMC) sampling process, equivalent to iteratively decoding and encoding, which allows us to sample from the learned latent distribution. Since, the generative model learns to map from the learned latent distribution, rather than the prior, we may use MCMC to improve the quality of samples drawn from the generative model, especially when the learned latent distribution is far from the prior. Using MCMC sampling, we are able to reveal previously unseen differences between generative autoencoders trained either with or without a denoising criterion.