# SCYNet: Testing supersymmetric models at the LHC with neural networks

## TL;DR

SCYNet is a neural network-based tool that rapidly evaluates the likelihood of supersymmetric models against LHC data, enabling efficient global fits and broader applicability to various new physics models.

## Contribution

The paper introduces a neural network approach for fast, accurate evaluation of supersymmetric model likelihoods at the LHC, including a model-independent method for wider applicability.

## Key findings

- Neural networks can evaluate the profile likelihood ratio within milliseconds.
- The approach enables fast global fits of the pMSSM-11.
- The model-independent neural network predicts likelihoods for various new physics models.

## Abstract

SCYNet (SUSY Calculating Yield Net) is a tool for testing supersymmetric models against LHC data. It uses neural network regression for a fast evaluation of the profile likelihood ratio. Two neural network approaches have been developed: one network has been trained using the parameters of the 11-dimensional phenomenological Minimal Supersymmetric Standard Model (pMSSM-11) as an input and evaluates the corresponding profile likelihood ratio within milliseconds. It can thus be used in global pMSSM-11 fits without time penalty. In the second approach, the neural network has been trained using model-independent signature-related objects, such as energies and particle multiplicities, which were estimated from the parameters of a given new physics model. While the calculation of the energies and particle multiplicities takes up computation time, the corresponding neural network is more general and can be used to predict the LHC profile likelihood ratio for a wider class of new physics models.