Is the Machine Smarter than the Theorist: Deriving Formulas for Particle Kinematics with Symbolic Regression

Abstract

We demonstrate the use of symbolic regression in deriving analytical formulas, which are needed at various stages of a typical experimental analysis in collider phenomenology. As a first application, we consider kinematic variables like the stransverse mass, $M_{T2}$, which are defined algorithmically through an optimization procedure and not in terms of an analytical formula. We then train a symbolic regression and obtain the correct analytical expressions for all known special cases of $M_{T2}$ in the literature. As a second application, we reproduce the correct analytical expression for a next-to-leading order (NLO) kinematic distribution from data, which is simulated with a NLO event generator. Finally, we derive analytical approximations for the NLO kinematic distributions after detector simulation, for which no known analytical formulas currently exist.

arXiv comment15 pages, 13 figures, 8 tables

References in corpus (31)

• MadGraph 5 : Going Beyond
• Identifying Boosted Objects with N-subjettiness
• Measuring masses of semi-invisibly decaying particles pair produced at hadron colliders
• m_T2 : the truth behind the glamour
• Discovering physical concepts with neural networks
• N-Jettiness: An Inclusive Event Shape to Veto Jets
• Symbolic Regression in Materials Science
• Gluino Stransverse Mass
• Phenomenology of event shapes at hadron colliders
• Measuring superparticle masses at hadron collider using the transverse mass kink
• Using Subsystem MT2 for Complete Mass Determinations in Decay Chains with Missing Energy at Hadron Colliders
• Dark Matter at Colliders: a Model-Independent Approach
• A Bayesian machine scientist to aid in the solution of challenging scientific problems
• A Review of the Mass Measurement Techniques proposed for the Large Hadron Collider
• A New SUSY mass reconstruction method at the CERN LHC
• Bisection-based asymmetric MT2 computation: a higher precision calculator than existing symmetric methods
• MTGEN : Mass scale measurements in pair-production at colliders
• Guide to transverse projections and mass-constraining variables
• Mass Determination of New States at Hadron Colliders
• AI Poincaré 2.0: Machine Learning Conservation Laws from Differential Equations
• Response of the CALICE Si-W Electromagnetic Calorimeter Physics Prototype to Electrons
• Lower Limit on Dark Matter Production at the Large Hadron Collider
• WIMP Dark Matter at the International Linear Collider
• On-shell constrained $M_2$ variables with applications to mass measurements and topology disambiguation
• The stransverse mass, MT2, in special cases
• Improving the sensitivity of stop searches with on-shell constrained invariant mass variables
• Full one-loop corrections to neutralino pair production in e+e- annihilation
• OPTIMASS: A Package for the Minimization of Kinematic Mass Functions with Constraints
• SYMBA: Symbolic Computation of Squared Amplitudes in High Energy Physics with Machine Learning
• Construction of a Kinematic Variable Sensitive to the Mass of the Standard Model Higgs Boson in H->WW*->l l nu nu-bar using Symbolic Regression
• Monophoton signals in light gravitino production at $e^{+}{e}^{-}$ colliders

Cited by in corpus (6)

• Quantum Vision Transformers for Quark-Gluon Classification
• Inferring the Equation of State from Neutron Star Observables via Machine Learning
• Discovering Nuclear Models from Symbolic Machine Learning
• Machine Learning Unveils the Power Law of Finite-Volume Energy Shifts
• $M_{TN}$ is all you need: production of multiple semi-invisible resonances at hadron colliders
• DeeLeMa: Missing information search with Deep Learning for Mass estimation