Abstract:Gaussian processes provide probabilistic surrogates for various applications including classification, uncertainty quantification, and optimization. Using a gradient-enhanced covariance matrix can be beneficial since it provides a more accurate surrogate relative to its gradient-free counterpart. An acute problem for Gaussian processes, particularly those that use gradients, is the ill-conditioning of their covariance matrices. Several methods have been developed to address this problem for gradient-enhanced Gaussian processes but they have various drawbacks such as limiting the data that can be used, imposing a minimum distance between evaluation points in the parameter space, or constraining the hyperparameters. In this paper a new method is presented that applies a diagonal preconditioner to the covariance matrix along with a modest nugget to ensure that the condition number of the covariance matrix is bounded, while avoiding the drawbacks listed above. Optimization results for a gradient-enhanced Bayesian optimizer with the Gaussian kernel are compared with the use of the new method, a baseline method that constrains the hyperparameters, and a rescaling method that increases the distance between evaluation points. The Bayesian optimizer with the new method converges the optimality, ie the $\ell_2$ norm of the gradient, an additional 5 to 9 orders of magnitude relative to when the baseline method is used and it does so in fewer iterations than with the rescaling method. The new method is available in the open source python library GpGradPy, which can be found at <a class="link-external link-https" href="https://github.com/marchildon/gpgradpy/tree/paper_precon" rel="external noopener nofollow">this https URL</a>. All of the figures in this paper can be reproduced with this library.

Gaussian Mass Optimization for Kernel Pca Parameters

The Optimization for Parameter of Kernel Function with GA

Generalized Convexity-Based Inexact Projection Method for Multiple Kernel Learning

Cross-validatory Framework for Optimal Parameter Estimation of KPCA and KPLS Models

A Kernel Function Optimization And Selection Algorithm Based On Cost Function Maximization

Gaussian kernel optimization: Complex problem and a simple solution

Stochastic Optimization for Kernel Pca

A Solution to the Ill-Conditioning of Gradient-Enhanced Covariance Matrices for Gaussian Processes

Kernel optimization-based discriminant analysis for face recognition

Global Optimization of Gaussian Process Acquisition Functions Using a Piecewise-Linear Kernel Approximation

A New Method for Determining the Parameter of Gaussian Kernel

Manifold Optimization Assisted Gaussian Variational Approximation

An Efficient Parameter Optimization of Maximum Correntropy Criterion.

Optimal Gaussian Kernel Parameter Selection for SVM Classifier

Multi-kernel PCA based high-dimensional images feature reduction

Kernel-Optimized Based Machine for Image Recognition

Optimizing Kernel Pca Using Sparse Representation-Based Classifier for Mstar Sar Image Target Recognition

Sparse Optimization for Second Order Kernel Methods

Hybrid kernel approach to improving the numerical stability of machine learning for parametric equations with Gaussian processes in the noisy and noise-free data assumptions

Efficient Model Selection for Support Vector Machine with Gaussian Kernel Function