Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo
The current work introduces a novel combination of two Bayesian tools, Gaussian Processes (GPs), and the use of the Approximate Bayesian Computation (ABC) algorithm for kernel selection and parameter estimation for machine learning applications. The combined methodology that this research article pr...
| Published in: | Frontiers in Built Environment |
|---|---|
| Main Authors: | , , , |
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2017-08-01
|
| Subjects: | |
| Online Access: | http://journal.frontiersin.org/article/10.3389/fbuil.2017.00052/full |
| _version_ | 1852725868923715584 |
|---|---|
| author | Anis Ben Abdessalem Nikolaos Dervilis David J. Wagg Keith Worden |
| author_facet | Anis Ben Abdessalem Nikolaos Dervilis David J. Wagg Keith Worden |
| author_sort | Anis Ben Abdessalem |
| collection | DOAJ |
| container_title | Frontiers in Built Environment |
| description | The current work introduces a novel combination of two Bayesian tools, Gaussian Processes (GPs), and the use of the Approximate Bayesian Computation (ABC) algorithm for kernel selection and parameter estimation for machine learning applications. The combined methodology that this research article proposes and investigates offers the possibility to use different metrics and summary statistics of the kernels used for Bayesian regression. The presented work moves a step toward online, robust, consistent, and automated mechanism to formulate optimal kernels (or even mean functions) and their hyperparameters simultaneously offering confidence evaluation when these tools are used for mathematical or engineering problems such as structural health monitoring (SHM) and system identification (SI). |
| format | Article |
| id | doaj-art-c1395490a7d14dcf92cfda947c2fc5de |
| institution | Directory of Open Access Journals |
| issn | 2297-3362 |
| language | English |
| publishDate | 2017-08-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| spelling | doaj-art-c1395490a7d14dcf92cfda947c2fc5de2025-08-19T21:10:33ZengFrontiers Media S.A.Frontiers in Built Environment2297-33622017-08-01310.3389/fbuil.2017.00052285105Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte CarloAnis Ben Abdessalem0Nikolaos Dervilis1David J. Wagg2Keith Worden3Dynamics Research Group, Department of Mechanical Engineering, University of Sheffield, Sheffield, United KingdomDynamics Research Group, Department of Mechanical Engineering, University of Sheffield, Sheffield, United KingdomDynamics Research Group, Department of Mechanical Engineering, University of Sheffield, Sheffield, United KingdomDynamics Research Group, Department of Mechanical Engineering, University of Sheffield, Sheffield, United KingdomThe current work introduces a novel combination of two Bayesian tools, Gaussian Processes (GPs), and the use of the Approximate Bayesian Computation (ABC) algorithm for kernel selection and parameter estimation for machine learning applications. The combined methodology that this research article proposes and investigates offers the possibility to use different metrics and summary statistics of the kernels used for Bayesian regression. The presented work moves a step toward online, robust, consistent, and automated mechanism to formulate optimal kernels (or even mean functions) and their hyperparameters simultaneously offering confidence evaluation when these tools are used for mathematical or engineering problems such as structural health monitoring (SHM) and system identification (SI).http://journal.frontiersin.org/article/10.3389/fbuil.2017.00052/fullkernel selectionhyperparameter estimationapproximate Bayesian computationsequential Monte CarloGaussian processes |
| spellingShingle | Anis Ben Abdessalem Nikolaos Dervilis David J. Wagg Keith Worden Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo kernel selection hyperparameter estimation approximate Bayesian computation sequential Monte Carlo Gaussian processes |
| title | Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo |
| title_full | Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo |
| title_fullStr | Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo |
| title_full_unstemmed | Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo |
| title_short | Automatic Kernel Selection for Gaussian Processes Regression with Approximate Bayesian Computation and Sequential Monte Carlo |
| title_sort | automatic kernel selection for gaussian processes regression with approximate bayesian computation and sequential monte carlo |
| topic | kernel selection hyperparameter estimation approximate Bayesian computation sequential Monte Carlo Gaussian processes |
| url | http://journal.frontiersin.org/article/10.3389/fbuil.2017.00052/full |
| work_keys_str_mv | AT anisbenabdessalem automatickernelselectionforgaussianprocessesregressionwithapproximatebayesiancomputationandsequentialmontecarlo AT nikolaosdervilis automatickernelselectionforgaussianprocessesregressionwithapproximatebayesiancomputationandsequentialmontecarlo AT davidjwagg automatickernelselectionforgaussianprocessesregressionwithapproximatebayesiancomputationandsequentialmontecarlo AT keithworden automatickernelselectionforgaussianprocessesregressionwithapproximatebayesiancomputationandsequentialmontecarlo |