Automatic differentiation for error analysis of Monte Carlo data

Abstract

Automatic Differentiation (AD) allows to determine exactly the Taylor series of any function truncated at any order. Here we propose to use AD techniques for Monte Carlo data analysis. We discuss how to estimate errors of a general function of measured observables in different Monte Carlo simulations. Our proposal combines the $\Gamma$-method with Automatic differentiation, allowing exact error propagation in arbitrary observables, even those defined via iterative algorithms. The case of special interest where we estimate the error in fit parameters is discussed in detail. We also present a freely available fortran reference implementation of the ideas discussed in this work.

arXiv commentLaTeX, 25 pags. More explicit notation. Software available in https://gitlab.ift.uam-csic.es/alberto/aderrors

References in corpus (4)

• Monte Carlo errors with less errors
• Critical slowing down and error analysis in lattice QCD simulations
• Critical slowing down of topological modes
• A Python program for the implementation of the Γ-method for Monte Carlo simulations

Cited by in corpus (28)

• Window observable for the hadronic vacuum polarization contribution to the muon $g-2$ from lattice QCD
• An analysis of systematic effects in finite size scaling studies using the gradient flow
• High precision renormalization of the flavour non-singlet Noether currents in lattice QCD with Wilson quarks
• Nuclear matrix elements from lattice QCD for electroweak and beyond-Standard-Model processes
• pyerrors: a python framework for error analysis of Monte Carlo data
• Scale setting and the light baryon spectrum in $N_f=2+1$ QCD with Wilson fermions
• Topological sampling through windings
• The gradient flow coupling at high-energy and the scale of SU(3) Yang-Mills theory
• Lattice determinations of the strong coupling
• Determination of ${\alpha }_s(m_Z)$ by the non-perturbative decoupling method
• Light quark masses in N_f = 2+1 lattice QCD with Wilson fermions
• Sampling the lattice Nambu-Goto string using Continuous Normalizing Flows
• Controlling Chaos in Van Der Pol Dynamics Using Signal-Encoded Deep Learning
• Determination of the charm quark mass in lattice QCD with $2+1$ flavours on fine lattices
• The renormalised $\mathrm{O}(a)$ improved vector current in three-flavour lattice QCD with Wilson quarks
• Ratio of flavour non-singlet and singlet scalar density renormalisation parameters in $N_{\mathrm{f}}=3$ QCD with Wilson quarks
• Heavy Wilson Quarks and O($a$) Improvement: Nonperturbative Results for $b_{\mathrm{g}}$
• Quantum Entanglement is Quantum: ZZ Production at the LHC
• Higher-Order Corrections to Quantum Observables in $h\to W{W}^{\ast }$
• Quantum computing fidelity susceptibility using automatic differentiation
• Quark mass dependence of the low-lying charmed mesons at one loop in HH$\chi$PT
• S-wave flavor-singlet meson mixing in QCD with light and charm quarks
• Precision lattice calculation of the hadronic contribution to the running of the electroweak gauge couplings
• The determination of potential scales in 2+1 flavor QCD
• Lattice investigation of custodial two-Higgs-doublet model at weak quartic couplings
• Memory efficient finite volume schemes with twisted boundary conditions
• Hadronic decay of vector charmonium from the lattice
• $f_K/f_{\pi }$ in iso-symmetric QCD and the CKM matrix unitarity