Pyro: Deep Universal Probabilistic Programming
This provides a flexible tool for AI researchers to build and scale complex probabilistic models, though it is incremental as it builds on existing frameworks like PyTorch.
The authors introduced Pyro, a probabilistic programming language built on Python and PyTorch to develop advanced probabilistic models in AI, enabling scalability to large datasets and high-dimensional models through stochastic variational inference and customizable behavior via Poutine.
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.