How to use Machine Learning to improve the discrimination between signal and background at particle colliders
Xabier Cid Vidal et al.
The popularity of Machine Learning (ML) has been increasing in the last decades in almost every area, being the commercial and scientific fields the most notorious ones. Concerning particle physics, ML hasbeen proved as a useful resource to make the most of projects such as the Large Hadron Collider (LHC). This means the way experiments work is changing to a new paradigm, in which smarter libraries areneeded to analyze the large amounts of data these experiments collect. The main goal is reducing the timeand effort put into the necessary calculations and predictions, while improving the performance. Withthis work we aim to encourage scientists at particle colliders to use ML and to try the different alternativeswe have available nowadays, focusing in the discrimination between signal and background. We assesssome of the most used libraries in the field, like Toolkit for Multivariate Data Analysis with ROOT,and also newer and more sophisticated options like PyTorch and Keras. We compare the performanceof different algorithms in simulated LHC data and produce some guidelines to help analysts deal withdifferent situations. Examples are the use of low or high level features from particle detectors or theamount of statistics available for training the algorithms.