New Developments on Computer Experiments with Quantitative and Qualitative Inputs
Loading...
Authors
Shahrokhian, Anita
Date
2025-01-30
Type
thesis
Language
eng
Keyword
Active Learning , Gaussian Process , BigData , Sequential Designs , Inverse Problem
Alternative Title
Abstract
Computer experiments with quantitative and qualitative inputs are widely used to study many scientific and engineering processes. Design, modeling and analysis of such computer experiments through emulators or surrogate models are important in the statistical literature. This thesis is devoted to three research topics on computer experiments with mixed inputs.
First, we propose a novel adaptive design approach for contour estimations in computer experiments with mixed inputs. By incorporating an adaptive search region strategy, the methodology improves contour estimation accuracy, supported by numerical studies and a real application. Second, we address the computational challenges in modeling large-scale computer experiments with mixed inputs for prediction. We extend existing methodologies, such as the Vecchia approximation, global-local Gaussian process, and local approximation Gaussian process, to accommodate both quantitative and qualitative inputs. Our proposed approach based on Vecchia approximation significantly improves computational efficiency and predictive accuracy in computer experiments with mixed inputs. Lastly, we introduce an adaptive design framework to enhance prediction accuracy in computer experiments with mixed inputs using a new decision criterion. This framework allows adaptive designs to update the model when the new observation changes the model accuracy significantly. This strategy helps the computational efficiency of Gaussian process with mixed inputs. The methodologies presented in this thesis provide practical tools and significant advancements for analyzing computer experiments with mixed inputs, offering new insights and opportunities for future research in this field.
Description
Citation
Publisher
License
Queen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.