Pyro: Deep Universal Probabilistic Programming

TL;DR

Pyro is a flexible probabilistic programming language built on Python and PyTorch, enabling scalable AI models with advanced inference algorithms and customizable components for complex probabilistic modeling.

Contribution

Pyro introduces a novel probabilistic programming framework that combines deep learning scalability with modular, customizable inference and modeling capabilities.

Findings

Supports large datasets and high-dimensional models

Utilizes stochastic variational inference algorithms

Provides composable building blocks for model customization

Abstract

Pyro is a probabilistic programming language built on Python as a platform for developing advanced probabilistic models in AI research. To scale to large datasets and high-dimensional models, Pyro uses stochastic variational inference algorithms and probability distributions built on top of PyTorch, a modern GPU-accelerated deep learning framework. To accommodate complex or model-specific algorithmic behavior, Pyro leverages Poutine, a library of composable building blocks for modifying the behavior of probabilistic programs.