.png){kind=icon}
Return
Influence functions for Gaussian process hyperparameter estimation
.png){kind=icon}
OA
AI
DOI：10.1007/s41060-026-01285-5.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We derive and analyse the influence function of hyperparameter estimators in Gaussian process (GP) models. While GPs are widely used for regression and probabilistic modelling, their hyperparameters are typically learned by maximizing the marginal likelihood. We show that the resulting estimators have unbounded influence functions, implying that even a single outlying data point can exert arbitrarily large leverage on kernel parameters. Using both theoretical derivations and empirical validations, we demonstrate that this lack of robustness is a structural property of GP hyperparameter estimation rather than a numerical artefact.
Keywords:
Gaussian processes
Influence functions
Hyperparameter estimation
Robust statistics
Robust regression
Journal
I
IF:
2.8
Papers:
1.1K
Citations:
1.3K
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveShareSaveShareSaveShareSave
View more