Dr. Juan Camilo Orduz · Sep 1, 2025

PyData Berlin 2025: Introduction to Stochastic Variational Inference with NumPyro

0Sign in to vote or save

This page cannot be shown here. You can still read it on the original site — the toolbar below keeps your place in the directory.

In this notebook we provide a brief introduction to Stochastic Variational Inference (SVI) with NumPyro . We provide the key mathematical concepts, but we focus on the code implementation. This introductory notebook is meant for practitioners. We do this by working through two examples: a very simple parameter recovery model and a Bayesian Neural Network. 
 This work was presented at PyData…

In this notebook we provide a brief introduction to Stochastic Variational Inference (SVI) with NumPyro. We provide the key mathematical concepts, but we focus on the code implementation. This introductory notebook is meant for practitioners. We do this by working through two examples: a very simple parameter recovery model and a Bayesian Neural Network.

This work was presented at PyData Berlin 2025, you can find the slides here.

Overview

Stochastic Variational Inference (SVI) is a scalable approximate inference method that transforms the problem of posterior inference into an optimization problem. Instead of sampling from the posterior distribution (like MCMC), SVI finds the best approximation to the posterior within a family of simpler distributions.

Read on /intro_svi/ ↗

Comments

Nothing yet. Say the first thing.