Deep Neural Networks for Physics Analysis on low-level whole-detector data at the LHC

TL;DR

This paper demonstrates the application of deep convolutional neural networks to entire low-level detector data, including calorimeter and track information, for physics analysis at the LHC, enabling direct event classification of background versus new physics signals.

Contribution

It introduces a novel approach using CNNs on full detector data, including sparse high-resolution images, for direct physics analysis, expanding beyond previous subset-focused methods.

Findings

CNNs can classify physics events using full detector data.

GPU and KNL architectures effectively process high-resolution sparse images.

The approach enables direct discrimination of background and new physics signals.

Abstract

There has been considerable recent activity applying deep convolutional neural nets (CNNs) to data from particle physics experiments. Current approaches on ATLAS/CMS have largely focussed on a subset of the calorimeter, and for identifying objects or particular particle types. We explore approaches that use the entire calorimeter, combined with track information, for directly conducting physics analyses: i.e. classifying events as known-physics background or new-physics signals. We use an existing RPV-Supersymmetry analysis as a case study and explore CNNs on multi-channel, high-resolution sparse images: applied on GPU and multi-node CPU architectures (including Knights Landing (KNL) Xeon Phi nodes) on the Cori supercomputer at NERSC.