Gaussian process hyper-parameter estimation using parallel asymptotically independent Markov sampling
From MaRDI portal
Abstract: Gaussian process emulators of computationally expensive computer codes provide fast statistical approximations to model physical processes. The training of these surrogates depends on the set of design points chosen to run the simulator. Due to computational cost, such training set is bound to be limited and quantifying the resulting uncertainty in the hyper-parameters of the emulator by uni-modal distributions is likely to induce bias. In order to quantify this uncertainty, this paper proposes a computationally efficient sampler based on an extension of Asymptotically Independent Markov Sampling, a recently developed algorithm for Bayesian inference. Structural uncertainty of the emulator is obtained as a by-product of the Bayesian treatment of the hyper-parameters. Additionally, the user can choose to perform stochastic optimisation to sample from a neighbourhood of the Maximum a Posteriori estimate, even in the presence of multimodality. Model uncertainty is also acknowledged through numerical stabilisation measures by including a nugget term in the formulation of the probability model. The efficiency of the proposed sampler is illustrated in examples where multi-modal distributions are encountered. For the purpose of reproducibility, further development, and use in other applications the code used to generate the examples is freely available for download at https://github.com/agarbuno/paims_codes
Recommendations
- Computer emulation with nonstationary Gaussian processes
- Robust Gaussian stochastic process emulation
- Cross-validation-based adaptive sampling for Gaussian process models
- Adaptive numerical designs for the calibration of computer codes
- Cholesky-based experimental design for Gaussian process and kernel-based emulation and calibration
Cites work
- scientific article; zbMATH DE number 1222285 (Why is no real title available?)
- scientific article; zbMATH DE number 720675 (Why is no real title available?)
- scientific article; zbMATH DE number 1098876 (Why is no real title available?)
- scientific article; zbMATH DE number 2117879 (Why is no real title available?)
- scientific article; zbMATH DE number 272681 (Why is no real title available?)
- scientific article; zbMATH DE number 961607 (Why is no real title available?)
- An adaptive Metropolis algorithm
- An adaptive sequential Monte Carlo sampler
- An efficient framework for optimal robust stochastic system design using stochastic simulation
- Bayesian calibration of computer models. (With discussion)
- Default priors for Gaussioan processes
- Galaxy formation: a Bayesian uncertainty analysis
- Gaussian processes for machine learning.
- Integrated likelihood methods for eliminating nuisance parameters. (With comments and a rejoinder).
- MCMC using Hamiltonian dynamics
- Monte Carlo strategies in scientific computing.
- Objective Bayesian analysis of spatial data with measurement error
- On Metropolis-Hastings algorithms with delayed rejection
- On adaptive resampling strategies for sequential Monte Carlo methods
- Particle Markov Chain Monte Carlo Methods
- Sequential Monte Carlo Samplers
- Slice sampling. (With discussions and rejoinder)
- The effect of the nugget on Gaussian process emulators of computer models
- The formal definition of reference priors
Cited in
(10)- Parallel cross-validation: a scalable fitting method for Gaussian process models
- Bayesian updating and model class selection with subset simulation
- A matrix-free approach for solving the parametric Gaussian process maximum likelihood problem
- scientific article; zbMATH DE number 7164769 (Why is no real title available?)
- Efficient optimization of the likelihood function in Gaussian process modelling
- Cholesky-based experimental design for Gaussian process and kernel-based emulation and calibration
- Accelerating pseudo-marginal MCMC using Gaussian processes
- Cross-validation-based adaptive sampling for Gaussian process models
- Subset simulation for probabilistic computer models
- Parallel partial Gaussian process emulation for computer models with massive output
This page was built for publication: Gaussian process hyper-parameter estimation using parallel asymptotically independent Markov sampling
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1659011)
