A program for the Bayesian Neural Network in the ROOT framework

TL;DR

This paper introduces a Bayesian Neural Network implementation in the ROOT framework's TMVA package, enhancing regression and probability fitting with uncertainty estimation and additional functionalities, demonstrated in High Energy Physics applications.

Contribution

The paper presents a novel Bayesian Neural Network algorithm integrated into ROOT's TMVA, offering improved regression, probability fitting, and uncertainty estimation capabilities.

Findings

Enhanced non-parametric regression and probability fitting.

Functionalities include cost function selection and complexity control.

Demonstrated application in High Energy Physics.

Abstract

We present a Bayesian Neural Network algorithm implemented in the TMVA package, within the ROOT framework. Comparing to the conventional utilization of Neural Network as discriminator, this new implementation has more advantages as a non-parametric regression tool, particularly for fitting probabilities. It provides functionalities including cost function selection, complexity control and uncertainty estimation. An example of such application in High Energy Physics is shown. The algorithm is available with ROOT release later than 5.29.