Towards new cross-validation-based estimators for Gaussian process regression: efficient adjoint computation of gradients
This work addresses parameter estimation in Gaussian process regression, offering incremental improvements in computational efficiency for researchers and practitioners in machine learning.
The authors tackled the problem of estimating covariance function parameters in Gaussian process regression by proposing new cross-validation criteria from scoring rules and an efficient gradient computation method, achieving lower complexity for evaluating leave-one-out criteria and gradients.
We consider the problem of estimating the parameters of the covariance function of a Gaussian process by cross-validation. We suggest using new cross-validation criteria derived from the literature of scoring rules. We also provide an efficient method for computing the gradient of a cross-validation criterion. To the best of our knowledge, our method is more efficient than what has been proposed in the literature so far. It makes it possible to lower the complexity of jointly evaluating leave-one-out criteria and their gradients.