Unbiased Gradient Estimation for Variational Auto-Encoders using Coupled Markov Chains
arXiv:2010.01845
Abstract
The variational auto-encoder (VAE) is a deep latent variable model that has two neural networks in an autoencoder-like architecture; one of them parameterizes the model's likelihood. Fitting its parameters via maximum likelihood (ML) is challenging since the computation of the marginal likelihood involves an intractable integral over the latent space; thus the VAE is trained instead by maximizing a variational lower bound. Here, we develop a ML training scheme for VAEs by introducing unbiased estimators of the log-likelihood gradient. We obtain the estimators by augmenting the latent space with a set of importance samples, similarly to the importance weighted auto-encoder (IWAE), and then constructing a Markov chain Monte Carlo coupling procedure on this augmented space. We provide the conditions under which the estimators can be computed in finite time and with finite variance. We show experimentally that VAEs fitted with unbiased estimators exhibit better predictive performance.
References in corpus (9)
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- Improving Variational Auto-Encoders using Householder Flow
- Importance Weighting and Variational Inference
- Unbiased Implicit Variational Inference
- Improving Variational Auto-Encoders using convex combination linear Inverse Autoregressive Flow
- Unbiased Smoothing using Particle Independent Metropolis-Hastings
- Reweighted Expectation Maximization
- SUMO: Unbiased Estimation of Log Marginal Probability for Latent Variable Models
- Double Happiness: Enhancing the Coupled Gains of L-lag Coupling via Control Variates