Higgs Boson Classification: Brain-inspired BCPNN Learning with StreamBrain

TL;DR

This paper explores brain-inspired BCPNN learning implemented in StreamBrain for high-energy physics data analysis, demonstrating competitive accuracy and AUC in Higgs Boson classification using HPC resources.

Contribution

It introduces StreamBrain, an HPC-optimized brain-inspired ML framework, and applies it to Higgs Boson data, showcasing its computational efficiency and classification performance.

Findings

Achieved up to 69.15% accuracy in Higgs Boson classification.

Reached 76.4% AUC performance on the dataset.

Demonstrated suitability of brain-inspired ML for HPC environments.

Abstract

One of the most promising approaches for data analysis and exploration of large data sets is Machine Learning techniques that are inspired by brain models. Such methods use alternative learning rules potentially more efficiently than established learning rules. In this work, we focus on the potential of brain-inspired ML for exploiting High-Performance Computing (HPC) resources to solve ML problems: we discuss the BCPNN and an HPC implementation, called StreamBrain, its computational cost, suitability to HPC systems. As an example, we use StreamBrain to analyze the Higgs Boson dataset from High Energy Physics and discriminate between background and signal classes in collisions of high-energy particle colliders. Overall, we reach up to 69.15% accuracy and 76.4% Area Under the Curve (AUC) performance.