“…The gradient‐free covariance matrix is commonly given by
where the hyperparameter
is the variance of the stationary residual error and
is a hyperparameter that estimates
, which is the true noise variance
34 . The hyperparameter
is used when the function evaluations are noisy and, in practice, it also serves to regularize
in order to reduce its condition number
34 . To separate the need to regularize the covariance matrix from the estimation of the uncertainty of the function evaluations, we use the following notation
…”