Automatic Differentiation Variational Inference

TL;DR

ADVI automates the derivation of variational inference algorithms for complex probabilistic models, enabling efficient Bayesian inference without manual derivation, thus accelerating model development and refinement.

Contribution

The paper introduces ADVI, a method that automatically derives variational inference algorithms for a broad class of models, removing the need for model-specific derivations.

Findings

ADVI supports a wide range of models without conjugacy.

ADVI scales to datasets with millions of observations.

ADVI is integrated into the Stan probabilistic programming system.

Abstract

Probabilistic modeling is iterative. A scientist posits a simple model, fits it to her data, refines it according to her analysis, and repeats. However, fitting complex models to large data is a bottleneck in this process. Deriving algorithms for new models can be both mathematically and computationally challenging, which makes it difficult to efficiently cycle through the steps. To this end, we develop automatic differentiation variational inference (ADVI). Using our method, the scientist only provides a probabilistic model and a dataset, nothing else. ADVI automatically derives an efficient variational inference algorithm, freeing the scientist to refine and explore many models. ADVI supports a broad class of models-no conjugacy assumptions are required. We study ADVI across ten different models and apply it to a dataset with millions of observations. ADVI is integrated into Stan, a probabilistic programming system; it is available for immediate use.