Equivariant neural networks for robust observables
arXiv:2405.13524 · doi:10.1103/PhysRevD.110.096023
Abstract
We introduce the usage of equivariant neural networks in the search for violations of the charge-parity () symmetry in particle interactions at the CERN Large Hadron Collider. We design neural networks that take as inputs kinematic information of recorded events and that transform equivariantly under the a symmetry group related to the transformation. We show that this algorithm allows to define observables reflecting the properties of the symmetry, showcasing its performance in several reference processes in top quark and electroweak physics. Imposing equivariance as an inductive bias in the algorithm improves the numerical convergence properties with respect to other methods that do not rely on equivariance and allows to construct optimal observables that significantly improve the state-of-the-art methodology in the searches considered.
Replacing with published version
References in corpus (34)
- The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
- The anti-k_t jet clustering algorithm
- An Introduction to PYTHIA 8.2
- DELPHES 3, A modular framework for fast simulation of a generic collider experiment
- Dimension-Six Terms in the Standard Model Lagrangian
- Renormalization Group Evolution of the Standard Model Dimension Six Operators III: Gauge Coupling Dependence and Phenomenology
- Mining gold from implicit models to improve likelihood-free inference
- Constraining Effective Field Theories with Machine Learning
- A Guide to Constraining Effective Field Theories with Machine Learning
- An Efficient Lorentz Equivariant Graph Neural Network for Jet Tagging
- SMEFTsim 3.0 -- a practical guide
- Automation of the matrix element reweighting method
- Measurement of the top quark polarization and spin correlations using dilepton final states in proton-proton collisions at 13 TeV
- Measurement of normalised multi-differential cross sections in pp collisions at TeV, and simultaneous determination of the strong coupling strength, top quark pole mass, and parton distribution functions
- Diboson Interference Resurrection
- On the maximal use of Monte Carlo samples: re-weighting events at NLO accuracy
- A set of top quark spin correlation and polarization observables for the LHC: Standard Model predictions and new physics contributions
- Analytical solution of ttbar dilepton equations
- EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets
- Equivariant Energy Flow Networks for Jet Tagging
- Optimized probes of -odd effects in the process at hadron colliders
- Benchmarking simplified template cross sections in production
- Measurement of the inclusive and differential WZ production cross sections, polarization angles, and triple gauge couplings in pp collisions at 13 TeV
- Parametrized classifiers for optimal EFT sensitivity
- Explainable Equivariant Neural Networks for Particle Physics: PELICAN
- Unbinned multivariate observables for global SMEFT analyses from machine learning
- Machine-enhanced CP-asymmetries in the Higgs sector
- Back to the Formula -- LHC Edition
- Measurement of the inclusive and differential cross sections in the dilepton channel and effective field theory interpretation in proton-proton collisions at 13 TeV
- Applications of Lattice Gauge Equivariant Neural Networks
- Measurement of the top quark pole mass using +jet events in the dilepton final state in proton-proton collisions at = 13 TeV
- Search for CP violation using events in the lepton+jets channel in pp collisions at = 13 TeV
- Equivariance and generalization in neural networks
- A rotation-equivariant graph neural network for learning hadronic SMEFT effects