Bayesian averaging of computer models with domain discrepancies: a nuclear physics perspective
arXiv:1904.04793
Abstract
This article studies Bayesian model averaging (BMA) in the context of competing expensive computer models in a typical nuclear physics setup. While it is well known that BMA accounts for the additional uncertainty of the model itself, we show that it also decreases the posterior variance of the prediction errors via an explicit decomposition. We extend BMA to the situation where the competing models are defined on non-identical study regions. Any model's local forecasting difficulty is offset by predictions obtained from the average model, thus extending individual models to the full domain. We illustrate our methodology via pedagogical simulations and applications to forecasting nuclear observables, which exhibit convincing improvements in both the BMA prediction error and empirical coverage probabilities.
References in corpus (9)
- Nuclear Energy Density Optimization
- Nuclear energy density optimization: Large deformations
- Error Estimates of Theoretical Models: a Guide
- Electric dipole polarizability and the neutron skin
- Nuclear energy density optimization: Shell structure
- Neutron drip line in the Ca region from Bayesian model averaging
- Uncertainty Quantification for Nuclear Density Functional Theory and Information Content of New Measurements
- The neutron drip line: single-particle degrees of freedom and pairing properties as sources of theoretical uncertainties
- Alpha Decay Energies of Superheavy Nuclei: Systematic Trends
Cited by in corpus (6)
- A Guided Tour of Ab Initio Nuclear Many-Body Theory
- Quantified limits of the nuclear landscape
- Get on the BAND Wagon: A Bayesian Framework for Quantifying Model Uncertainties in Nuclear Dynamics
- Beyond the proton drip line: Bayesian analysis of proton-emitting nuclei
- Statistical aspects of nuclear mass models
- Lightweight self-conjugate nucleus Zr