Improved information criteria for Bayesian model averaging in lattice field theory

Abstract

Bayesian model averaging is a practical method for dealing with uncertainty due to model specification. Use of this technique requires the estimation of model probability weights. In this work, we revisit the derivation of estimators for these model weights. Use of the Kullback-Leibler divergence as a starting point leads naturally to a number of alternative information criteria suitable for Bayesian model weight estimation. We explore three such criteria, known to the statistics literature before, in detail: a Bayesian analogue of the Akaike information criterion which we call the BAIC, the Bayesian predictive information criterion (BPIC), and the posterior predictive information criterion (PPIC). We compare the use of these information criteria in numerical analysis problems common in lattice field theory calculations. We find that the PPIC has the most appealing theoretical properties and can give the best performance in terms of model-averaging uncertainty, particularly in the presence of noisy data, while the BAIC is a simple and reliable alternative.

70 pages, 13 figures. v2: corrections to data subset formulas for BPIC and PPIC; edits for clarity. v3: updated to journal version

References in corpus (15)

• Why your model parameter confidences might be too optimistic -- unbiased estimation of the inverse covariance matrix
• Ab initio calculation of the neutron-proton mass difference
• Nonlinear shrinkage estimation of large-dimensional covariance matrices
• Adaptive Multidimensional Integration: VEGAS Enhanced
• Bayesian model averaging for analysis of lattice field theory results
• Get on the BAND Wagon: A Bayesian Framework for Quantifying Model Uncertainties in Nuclear Dynamics
• Update: Accurate Determinations of alpha_s from Realistic Lattice QCD
• Bayesian Methods for Parameter Estimation in Effective Field Theories
• Lattice QCD determination of neutron-antineutron matrix elements with physical quark masses
• $F_K/F_{\pi }$ from Möbius domain-wall fermions solved on gradient-flowed HISQ ensembles
• Sign problems, noise, and chiral symmetry breaking in a QCD-like theory
• D-meson semileptonic decays to pseudoscalars from four-flavor lattice QCD
• Bayesian Model Averaging in Astrophysics: A Review
• Model averaging approaches to data subset selection
• Does Bayesian Model Averaging improve polynomial extrapolations? Two toy problems as tests

Cited by in corpus (11)

• Precision calculation of the electromagnetic radii of the proton and neutron from lattice QCD
• Electromagnetic form factors of the nucleon from $N_f=2+1$ lattice QCD
• Investigation of pion-nucleon contributions to nucleon matrix elements
• Deep learning of phase transitions with minimal examples
• Kaon radiative leptonic decay rates from lattice QCD simulations at the physical point
• A generalized statistical model for fits to parton distributions
• Zemach and Friar radii of the proton and neutron from lattice QCD
• The $\pi \pi$ scattering amplitude at large $N_{\text{c}}$
• Single Particle Spectrum of Doped ${\mathrm{C}}_{20}{\mathrm{H}}_{12}$-Perylene
• The determination of potential scales in 2+1 flavor QCD
• Proton and neutron electromagnetic form factors from lattice QCD in the continuum limit