Nonnegativity-enforced Gaussian process regression
ABSTRACT: Gaussian process (GP) regression is a flexible non-parametric approach to approximate complex models. In many cases, these models correspond to processes with bounded physical properties. Standard GP regression typically results in a proxy model which is unbounded for all temporal or spaci...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2020-03-01
|
| Series: | Theoretical and Applied Mechanics Letters |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2095034920300313 |
| _version_ | 1819265261554892800 |
|---|---|
| author | Andrew Pensoneault Xiu Yang Xueyu Zhu |
| author_facet | Andrew Pensoneault Xiu Yang Xueyu Zhu |
| author_sort | Andrew Pensoneault |
| collection | DOAJ |
| description | ABSTRACT: Gaussian process (GP) regression is a flexible non-parametric approach to approximate complex models. In many cases, these models correspond to processes with bounded physical properties. Standard GP regression typically results in a proxy model which is unbounded for all temporal or spacial points, and thus leaves the possibility of taking on infeasible values. We propose an approach to enforce the physical constraints in a probabilistic way under the GP regression framework. In addition, this new approach reduces the variance in the resulting GP model. |
| first_indexed | 2024-12-23T20:42:34Z |
| format | Article |
| id | doaj.art-b86481e87eda453a90e7b0655393c958 |
| institution | Directory Open Access Journal |
| issn | 2095-0349 |
| language | English |
| last_indexed | 2024-12-23T20:42:34Z |
| publishDate | 2020-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Theoretical and Applied Mechanics Letters |
| spelling | doaj.art-b86481e87eda453a90e7b0655393c9582022-12-21T17:31:52ZengElsevierTheoretical and Applied Mechanics Letters2095-03492020-03-01103182187Nonnegativity-enforced Gaussian process regressionAndrew Pensoneault0Xiu Yang1Xueyu Zhu2Department of Mathematics, University of Iowa, Iowa, IA 52246, USADepartment of Industrial and Systems Engineering, Lehigh University, Bethlehem, PA 18015, USA; Corresponding author. (X. Yang).Department of Mathematics, University of Iowa, Iowa, IA 52246, USA; Corresponding author. (X. Zhu).ABSTRACT: Gaussian process (GP) regression is a flexible non-parametric approach to approximate complex models. In many cases, these models correspond to processes with bounded physical properties. Standard GP regression typically results in a proxy model which is unbounded for all temporal or spacial points, and thus leaves the possibility of taking on infeasible values. We propose an approach to enforce the physical constraints in a probabilistic way under the GP regression framework. In addition, this new approach reduces the variance in the resulting GP model.http://www.sciencedirect.com/science/article/pii/S2095034920300313Gaussian process regressionConstrained optimization |
| spellingShingle | Andrew Pensoneault Xiu Yang Xueyu Zhu Nonnegativity-enforced Gaussian process regression Theoretical and Applied Mechanics Letters Gaussian process regression Constrained optimization |
| title | Nonnegativity-enforced Gaussian process regression |
| title_full | Nonnegativity-enforced Gaussian process regression |
| title_fullStr | Nonnegativity-enforced Gaussian process regression |
| title_full_unstemmed | Nonnegativity-enforced Gaussian process regression |
| title_short | Nonnegativity-enforced Gaussian process regression |
| title_sort | nonnegativity enforced gaussian process regression |
| topic | Gaussian process regression Constrained optimization |
| url | http://www.sciencedirect.com/science/article/pii/S2095034920300313 |
| work_keys_str_mv | AT andrewpensoneault nonnegativityenforcedgaussianprocessregression AT xiuyang nonnegativityenforcedgaussianprocessregression AT xueyuzhu nonnegativityenforcedgaussianprocessregression |