Abstract
Gaussian process regression (GPR), is a powerful non-parametric approach for data modeling, which has garnered considerable interest in the past decade, however its widespread application is impeded by the significant computational burden for larger datasets. The computational complexity for both inference and hyperparameter learning in GPs lead to O(N3) for N training points. The current state-of-the-art approximations, such as structured kernel interpolation (SKI)-based methods e.g., Kernel Interpolation for Scalable Structured Gaussian Process (KISSGP), have emerged to mitigate this challenge by providing a scalable inducing point alternatives. However, the choice of the optimal number of grid points, which influences the accuracy and efficiency of the model, typically remains fixed and is chosen arbitrarily. In this work, we introduce a novel approximation framework, Malleable KISSGP (MKISSGP), which dynamically adjusts grid points using a new hyperparameter of the model called density, which adapts to the changes in the kernel hyperparameters in each training iteration. In comparison with the state-of-the-art KISSGP and irrespective of changes in hyperparameters, our proposed MKISSGP algorithm exhibits consistent error levels in the reconstruction of the kernel matrix, and offers reduced computational complexity. We present extensive simulations to validate the improved performance of the proposed MKISSGP, and give directions for future research.
Original language | English |
---|---|
Title of host publication | 32nd European Signal Processing Conference, EUSIPCO 2024 - Proceedings |
Publisher | European Signal Processing Conference, EUSIPCO |
Pages | 997-1001 |
Number of pages | 5 |
ISBN (Electronic) | 9789464593617 |
DOIs | |
Publication status | Published - 2024 |
Event | 32nd European Signal Processing Conference, EUSIPCO 2024 - Lyon, France Duration: 26 Aug 2024 → 30 Aug 2024 https://eusipcolyon.sciencesconf.org/ |
Publication series
Name | European Signal Processing Conference |
---|---|
ISSN (Print) | 2219-5491 |
Conference
Conference | 32nd European Signal Processing Conference, EUSIPCO 2024 |
---|---|
Abbreviated title | EUSIPCO 2024 |
Country/Territory | France |
City | Lyon |
Period | 26/08/24 → 30/08/24 |
Internet address |
Bibliographical note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-careOtherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Keywords
- Gaussian process regression
- KISSGP
- Low-rank approximation
- Structured kernel interpolation