Bayes Blocks: An Implementation of the Variational Bayesian Building Blocks Framework

TL;DR

Bayes Blocks is a user-friendly software library that simplifies constructing and learning complex probabilistic models using variational Bayesian methods, with versatile building blocks for various model types.

Contribution

It introduces a flexible, easy-to-use library with a comprehensive set of building blocks for probabilistic modeling based on variational Bayesian techniques.

Findings

Supports a wide range of static and dynamic models

Enables fast and robust variational Bayesian inference

Provides diverse variable types and computational nodes

Abstract

A software library for constructing and learning probabilistic models is presented. The library offers a set of building blocks from which a large variety of static and dynamic models can be built. These include hierarchical models for variances of other variables and many nonlinear models. The underlying variational Bayesian machinery, providing for fast and robust estimation but being mathematically rather involved, is almost completely hidden from the user thus making it very easy to use the library. The building blocks include Gaussian, rectified Gaussian and mixture-of-Gaussians variables and computational nodes which can be combined rather freely.