Optimal Variance Control of the Score Function Gradient Estimator for Importance Weighted Bounds

Abstract

This paper introduces novel results for the score function gradient estimator of the importance weighted variational bound (IWAE). We prove that in the limit of large $K$ (number of importance samples) one can choose the control variate such that the Signal-to-Noise ratio (SNR) of the estimator grows as $\sqrt{K}$. This is in contrast to the standard pathwise gradient estimator where the SNR decreases as $1/\sqrt{K}$. Based on our theoretical findings we develop a novel control variate that extends on VIMCO. Empirically, for the training of both continuous and discrete generative models, the proposed method yields superior variance reduction, resulting in an SNR for IWAE that increases with $K$ without relying on the reparameterization trick. The novel estimator is competitive with state-of-the-art reparameterization-free gradient estimators such as Reweighted Wake-Sleep (RWS) and the thermodynamic variational objective (TVO) when training generative models.

References in corpus (12)

• Auto-Encoding Variational Bayes
• Categorical Reparameterization with Gumbel-Softmax
• Neural Discrete Representation Learning
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Variational Lossy Autoencoder
• REBAR: Low-variance, unbiased gradient estimates for discrete latent variable models
• Variational inference for Monte Carlo objectives
• BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling
• Deep AutoRegressive Networks
• Backpropagation through the Void: Optimizing control variates for black-box gradient estimation