Software codes

CRPropa is a publicly available simulation framework to study the propagation of ultra-high-energy nuclei up to iron on their voyage through an (extra)galactic environment. It takes into account: pion production, photodisintegration and energy losses by pair production of all relevant isotopes in the ambient low-energy photon fields as well as nuclear decay.

The EPiC-GAN – equivariant point cloud generative adversarial network – generates permutation equivariant point clouds of variable cardinality. A first case study explored the generation of particle jets in hadronic coordinates (p_T, rapidity η, angle φ) with variable particle multiplicity. The EPiC-GANs were trained on the JetNet30 and the JetNet150 gluon, light quark, and top datasets, respectively.

HiggsTools is a toolbox for comparing a wide class of BSM models to all available experimental results from searches for new (scalar) particles and measurements of the 125GeV Higgs boson at colliders. HiggsTools is composed of three sub-libraries: HiggsPredictions, HiggsBounds, HiggsSignals.

A python toolkit for silicon photomultiplier (SiPM) simulation. This program can be used to simulate primary Geiger discharges and generate correlated Geiger discharges due to prompt and delayed cross-talk and after-pulses. In addition, charge spectra, current transients and time-difference distributions can be generated.

The MasterCode is a computer code that allows to fit different versions of the Minimal Supersymmetric Standard Model (MSSM) to currently existing experimental data.

OmniJet-α is the first multi-task foundation model for particle physics at the LHC, which employs innovative tokenization and transfer learning techniques to effectively encode jet data and bridge unsupervised and supervised tasks.

PeakOTron is a tool to perform automated fits of silicon photomultiplier (SiPM) charge spectra to extract useful parameters from experiments. PeakOTron can be used for detailed analyses of single spectra, and also for the automatic characterisation of large samples of SiPMs.

Anomaly detection is a class of machine learning methods that aims to find unexpected patterns in data. CATHODE is a powerful such method that is designed to find new physics processes with minimal model assumptions. The sk_cathode code base provides easy-to-use modules following a sklearn-like API for building and testing CATHODE and similar methods. It also provides a collection of jupyter notebook tutorials that guide through core concepts of anomaly detection in particle physics.