Hierarchical Variational Models

About

Black box variational inference allows researchers to easily prototype and evaluate an array of models. Recent advances allow such algorithms to scale to high dimensions. However, a central question remains: How to specify an expressive variational distribution that maintains efficient computation? To address this, we develop hierarchical variational models (HVMs). HVMs augment a variational approximation with a prior on its parameters, which allows it to capture complex structure for both discrete and continuous latent variables. The algorithm we develop is black box, can be used for any HVM, and has the same computational efficiency as the original approximation. We study HVMs on a variety of deep discrete latent variable models. HVMs generalize other expressive variational distributions and maintains higher fidelity to the posterior.

Rajesh Ranganath, Dustin Tran, David M. Blei · 2015

Related benchmarks

Task	Dataset	Result
Unsupervised Learning	CIFAR-10 Pinwheel prior	Latent NLL1.69	6
Unsupervised Learning	CIFAR-10 Square prior	Latent NLL1.64	6
Unsupervised Learning	MNIST Pinwheel prior (test)	MMD0.74	6
Unsupervised Learning	MNIST Square prior (test)	MMD0.76	6
Unsupervised Learning	CIFAR-10 Swiss Roll prior	Latent NLL5.74	6
Unsupervised Learning	MNIST Swiss Roll prior (test)	MMD0.7	6