Resampled Priors for Variational Autoencoders
arXiv:1810.11428
Abstract
We propose Learned Accept/Reject Sampling (LARS), a method for constructing richer priors using rejection sampling with a learned acceptance function. This work is motivated by recent analyses of the VAE objective, which pointed out that commonly used simple priors can lead to underfitting. As the distribution induced by LARS involves an intractable normalizing constant, we show how to estimate it and its gradients efficiently. We demonstrate that LARS priors improve VAE performance on several standard datasets both when they are learned jointly with the rest of the model and when they are fitted to a pretrained model. Finally, we show that LARS can be combined with existing methods for defining flexible priors for an additional boost in performance.
References in corpus (1)
Cited by in corpus (6)
- Generative Latent Flow
- On the Necessity and Effectiveness of Learning the Prior of Variational Auto-Encoder
- Variational Autoencoders with Riemannian Brownian Motion Priors
- Bigeminal Priors Variational auto-encoder
- Energy-Inspired Models: Learning with Sampler-Induced Distributions
- Characterizing and Avoiding Problematic Global Optima of Variational Autoencoders