eprintid: 10157434
rev_number: 7
eprint_status: archive
userid: 699
dir: disk0/10/15/74/34
datestamp: 2022-10-18 16:50:50
lastmod: 2022-10-18 16:50:50
status_changed: 2022-10-18 16:50:50
type: article
metadata_visibility: show
sword_depositor: 699
creators_name: Ming, Deyu
creators_name: Williamson, Daniel
creators_name: Guillas, Serge
title: Deep Gaussian Process Emulation using Stochastic Imputation
ispublished: inpress
divisions: C06
divisions: F61
divisions: B04
divisions: UCL
divisions: C05
divisions: F49
keywords: Elliptical slice sampling, Linked Gaussian processes, Option Greeks, Surrogate model, Stochastic expectation maximization
note: © 2022 The Author(s). Published with license by Taylor & Francis Group, LLC.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
abstract: Deep Gaussian processes (DGPs) provide a rich class of models that can better represent functions with varying regimes or sharp changes, compared to conventional GPs. In this work, we propose a novel inference method for DGPs for computer model emulation. By stochastically imputing the latent layers, our approach transforms a DGP into a linked GP: a novel emulator developed for systems of linked computer models. This transformation permits an efficient DGP training procedure that only involves optimizations of conventional GPs. In addition, predictions from DGP emulators can be made in a fast and analytically tractable manner by naturally using the closed form predictive means and variances of linked GP emulators. We demonstrate the method in a series of synthetic examples and empirical applications, and show that it is a competitive candidate for DGP surrogate inference, combining efficiency that is comparable to doubly stochastic variational inference and uncertainty quantification that is comparable to the fully-Bayesian approach. A Python package dgpsi implementing the method is also produced and available at https://github.com/mingdeyu/DGP.
date: 2022-10-12
date_type: published
publisher: Taylor & Francis
official_url: https://doi.org/10.1080/00401706.2022.2124311
oa_status: green
full_text_type: pub
language: eng
primo: open
primo_central: open_green
verified: verified_manual
elements_id: 1877530
doi: 10.1080/00401706.2022.2124311
lyricists_name: Guillas, Serge
lyricists_name: Ming, Deyu
lyricists_id: SGUIL73
lyricists_id: MINGX71
actors_name: Ming, Deyu
actors_id: MINGX71
actors_role: owner
full_text_status: public
publication: Technometrics
citation:        Ming, Deyu;    Williamson, Daniel;    Guillas, Serge;      (2022)    Deep Gaussian Process Emulation using Stochastic Imputation.                   Technometrics        10.1080/00401706.2022.2124311 <https://doi.org/10.1080/00401706.2022.2124311>.    (In press).    Green open access   
 
document_url: https://discovery.ucl.ac.uk/id/eprint/10157434/1/Deep%20Gaussian%20Process%20Emulation%20using%20Stochastic%20Imputation.pdf