|
Activity Number:
|
354
- Experimental Design and Reliability
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
|
Abstract #318602
|
|
|
Title:
|
A Simple Gaussian Process Modeling Approach for Experiments with Quantitative-Sequence Factors
|
|
Author(s):
|
Yanran Wei* and X D
|
|
Companies:
|
Virginia Tech and Department of Statistics, Virginia Polytechnic Institute and State University
|
|
Keywords:
|
Analysis of experimental data;
Gaussian process;
Permutation matrix;
QS factor
|
|
Abstract:
|
Experimental data with complex structures are emerging in the modern applications, such as healthcare and drug discovery. This work concerns the analysis of data from experiments of arranging several components in a sequence order associated with their quantities. We call such an experiment as the experiment with quantitative-sequence (QS) factor. The experimental data with QS can potentially involve a large number of data points with the increasing size of components. We proposed a simple Gaussian Process model for analyzing the experimental data with QS factor by transforming the QS factor into a generalized permutation matrix. The computation of the model estimation is facilitated by the use of the inducing points. The performance of the proposed method is evaluated through several numerical examples.
|
Authors who are presenting talks have a * after their name.
