Higgs Boson Classification

In particle physics, Higgs Boson to tau-tau decay signals are notoriously difficult to identify due to the presence of severe background noise generated by other decaying particles. This approach uses neural networks to classify events as signals or background noise.

Keywords: Machine Learning, Binary Classification, Neural Networks

Implementation

Paper

Poster

Usage

Full instructions coming soon.

Dataset

CERN dataset

TensorFlow on Amazon EC2 GPU Cluster

For GPU-accelerated training: Setup Script

Training

Train the model:

$ sudo python higgs_neural_net.py