Learning to see R-parity violating scalar top decays

TL;DR

This paper applies advanced computer vision transformer models to particle physics event classification, demonstrating improved sensitivity in detecting R-parity violating scalar top decays compared to traditional CNNs.

Contribution

It introduces transformer-based models like CoAtNet and MaxViT for jet substructure analysis in particle physics, showing their superior performance over CNNs in this context.

Findings

Transformer models outperform CNNs in classifying signal vs. background.

Replacing CNN with MaxViT nearly doubles the $S/\sqrt{B}$ ratio.

Using MaxViT enhances the exclusion limits for stop and neutralino masses.

Abstract

With this article we introduce recent, improved machine learning methods from computer vision to the problem of event classification in particle physics. Supersymmetric scalar top decays to top quarks and weak scale bino-like neutralinos, where the neutralinos decay via the $UDD$ operator to three quarks, are difficult to search for and therefore weakly constrained. The jet substructure of the boosted decay products can be used to differentiate signal from background events. We apply transformer-based computer vision models CoAtNet and MaxViT to images built from jet constituents and compare the classification performance to a more classical convolutional neural network (CNN). We find that results from computer vision translate well onto physics applications and both transformer-based models perform better than the CNN. By replacing the CNN with MaxViT we find an improvement of $S/\sqrt{B}$ by a factor of almost 2 for some neutralino masses. We show that combining this classifier with additional features results in a strong separation of background and signal. We also find that replacing a CNN with a MaxViT model in a simple mock analysis can push the 95% C.L. exclusion limit of stop masses by about $100$ GeV and $60$ GeV for neutralino masses of $100$ GeV and $500$ GeV.