Bring the noise: exact inference from noisy simulations in collider physics

Abstract

We rely on Monte Carlo (MC) simulations to interpret searches for new physics at the Large Hadron Collider (LHC) and elsewhere. These simulations result in noisy and approximate estimators of selection efficiencies and likelihoods. In this context we pioneer an exact-approximate computational method - exact-approximate Markov Chain Monte Carlo (MCMC), also known as pseudo-marginal MCMC - that returns exact inferences despite noisy simulations. To do so, we introduce an unbiased estimator for a Poisson likelihood. We demonstrate the new estimator and new techniques in examples based on a search for neutralinos and charginos at the LHC using a simplified model. We find attractive performance characteristics - exact inferences are obtained for a similar computational cost to approximate ones from existing methods and inferences are robust with respect to the number of events generated per point. The unbiased estimator uses a Poisson-distributed number of MC events; it is also possible to construct a biased estimator whose bias decays factorially with increasing number of MC events.

12 pages, 5 figures. closely matches version accepted for publication

References in corpus (25)

• emcee: The MCMC Hammer
• An Introduction to PYTHIA 8.2
• MCMC using Hamiltonian dynamics
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• Herwig++ Physics and Manual
• Rank-normalization, folding, and localization: An improved $\hat{R}$ for assessing convergence of MCMC
• Event Generation with Sherpa 2.2
• The pseudo-marginal approach for efficient Monte Carlo computations
• Precision predictions for electroweak superpartner production at hadron colliders with Resummino
• Gaugino production in proton-proton collisions at a center-of-mass energy of 8 TeV
• On the efficiency of pseudo-marginal random walk Metropolis algorithms
• GAMBIT: The Global and Modular Beyond-the-Standard-Model Inference Tool
• Search for chargino--neutralino pair production in final states with three leptons and missing transverse momentum in $\sqrt{s}=13$ TeV $pp$ collisions with the ATLAS detector
• On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods
• ColliderBit: a GAMBIT module for the calculation of high-energy collider observables and likelihoods
• Constraining new physics with SModelS version 2
• Bias in parametric estimation: reduction and useful side-effects
• Convergence properties of pseudo-marginal Markov chain Monte Carlo algorithms
• A Noisy Monte Carlo Algorithm
• A binned likelihood for stochastic models
• On nonnegative unbiased estimators
• Simple and statistically sound recommendations for analysing physical theories
• Probabilistic treatment of the uncertainty from the finite size of weighted Monte Carlo data
• A unified perspective on modified Poisson likelihoods for limited Monte Carlo data
• Under-coverage in high-statistics counting experiments with finite MC samples