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

TL;DR

This paper introduces a new diagonal preconditioning method with a modest nugget to stabilize gradient-enhanced Gaussian process covariance matrices, improving optimization convergence and avoiding prior drawbacks.

Contribution

A novel preconditioning approach for gradient-enhanced Gaussian processes that bounds the covariance matrix condition number without restricting data or hyperparameters.

Findings

Significantly improved convergence in Bayesian optimization.

Outperforms baseline and rescaling methods in fewer iterations.

Method is implemented in an open-source Python library.

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 https://github.com/marchildon/gpgradpy/tree/paper_precon. All of the figures in this paper can be reproduced with this library.