BayesFusion, LLC

03/21/2025

Hybrid Dynamic Bayesian Networks video is now available on our YouTube channel.

...

10/17/2024

SMILE 2.3 is now available. This version of the library supports equations in dynamic Bayesian networks.

The libraries for C++, Python, Java, R and .NET can be downloaded from https://download.bayesfusion.com. We also maintain repositories for use with Maven and pip. Please see the download website for more details.

Supported Python versions are 3.9 to 3.13.

10/13/2024

GeNIe 5.0, available for download as of this date, is our next big step toward full fusion of Bayesian networks with systems of equations. It allows for continuous nodes inside DBNs (Dynamic Bayesian Networks), specified by means of equations and continuous probability distributions. This allows for building models that mimic systems of difference equations, deriving insightful graphs such as temporal value plots or phase space diagrams. These models are solved not symbolically but rather by simulation, which allows for complete modeling freedom and exploration of dynamic systems with noise. Other new features include Most Probable Explanation (MPE) and bulk update.

12/12/2023

SMILE 2.2 is now available. This version of the library supports new diagnosis algorithms.

The libraries for C++, Python, Java, R and .NET can be downloaded from https://download.bayesfusion.com. We also maintain repositories for use with Maven and pip, see the download website for more details.

Supported Python versions are 3.8 to 3.12.

03/08/2023

BayesBox Dashboard provides at-a-glance views of selected key variables that are relevant to a given objective or a business process. Watch our new YouTube video:

is an online viewer/repository for Bayesian networks and other model types...

03/07/2023

BayesFusion’s model repository at https://repo.bayesfusion.com offers a collection of example models for our users. We encourage our users to contribute to the user community by making their models available. In case of models described in publications, it is a great way of advertising your work. An on-screen comment that lists the correct reference to the paper in which the model appeared should increase your citation count. Please contact us in case you want to contribute your model to the repository.

10/29/2022

Metalog Builder
BayesFusion has released Metalog Builder, an interactive tool for constructing metalog distributions, accessible to the decision modeling community through a web interface at https://metalog.bayesfusion.com/. Metalog distribution (also known as the Keelin distribution) is extremely flexible and is capable of fitting many naturally occurring distributions. It can be specified by probability quantiles and it is able to represent distributions that are unbounded, semi-bounded, and bounded.

10/24/2022

Probability Distribution Visualizer

BayesFusion has released Probability Distribution Visualizer, an interactive tool for visualizing expressions with probability distributions through Monte Carlo simulation. When the distribution is transformed by an equation expression, the task is daunting even for an experienced decision analyst and requires considerable statistical insight. The tool aids the process of choosing the right probability distribution over continuous data. It is accessible to the decision modeling community through a web interface at https://prob.bayesfusion.com/.

10/13/2022

PGM-2022 BAYESFUSION BEST STUDENT PAPER AWARD WINNER ANNOUNCED

The Program Committee of the 11th International Conference on Probabilistic Graphical Models (PGM 2022) announced the winner of the BayesFusion Best Student Paper Award in Almeria, Spain, on October 6. The winner is:

Enrico Giudice, Department of Mathematics and Computer Science, University of Basel, Switzerland, for the paper entitled The Dual PC Algorithm for Structure Learning, co-authored with Jack Kuipers and Giusi Moffa.

Our heartfelt congratulations!

07/15/2022

SMILE 2.0 is now available. This version of the library supports discrete node outcomes based on numeric intervals or point values. Also, the metalog probability distribution can be used in equation node definitions.
The libraries for C++, Python, Java, R and .NET can be downloaded from https://download.bayesfusion.com. We also maintain repositories for use with Maven and pip, see the download website for more details.

07/15/2022

Starting with version 2.0, SMILE fully supports Macs based on M series (Apple Silicon) ARM CPUs. The C++, Python and Java libraries are available as universal binaries, containing both ARM and x64 code. The wrapper for R is available as separate binaries for ARM and x64.

To download the libraries, go to

07/13/2022

GeNIe 4.0 is now available at https://lnkd.in/dZMHsdC8.
Most important new features are:
- discrete nodes with outcomes based on numeric intervals or point values
- metalog distribution, including interactive metalog builder tool
- geospatial processing added, Esri ASCII raster grids supported
- Distribution Visualizer window