Automated Variational Inference in Probabilistic Programming

Abstract

We present a new algorithm for approximate inference in probabilistic programs, based on a stochastic gradient for variational programs. This method is efficient without restrictions on the probabilistic program; it is particularly practical for distributions which are not analytically tractable, including highly structured distributions that arise in probabilistic programs. We show how to automatically derive mean-field probabilistic programs and optimize them, and demonstrate that our perspective improves inference efficiency over other algorithms.

References in corpus (4)

• Stochastic Variational Inference
• Hybrid Variational/Gibbs Collapsed Inference in Topic Models
• Optimization of Structured Mean Field Objectives
• Variational Program Inference

Cited by in corpus (1)

• Automatic Variational Inference in Stan