Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Bandwidth Selection for Gaussian Kernel Ridge Regression via Jacobian Control
(Submitted on 24 May 2022 (v1), last revised 1 Dec 2023 (this version, v4))
Abstract: Most machine learning methods require tuning of hyper-parameters. For kernel ridge regression with the Gaussian kernel, the hyper-parameter is the bandwidth. The bandwidth specifies the length scale of the kernel and has to be carefully selected to obtain a model with good generalization. The default methods for bandwidth selection, cross-validation and marginal likelihood maximization, often yield good results, albeit at high computational costs. Inspired by Jacobian regularization, we formulate an approximate expression for how the derivatives of the functions inferred by kernel ridge regression with the Gaussian kernel depend on the kernel bandwidth. We use this expression to propose a closed-form, computationally feather-light, bandwidth selection heuristic, based on controlling the Jacobian. In addition, the Jacobian expression illuminates how the bandwidth selection is a trade-off between the smoothness of the inferred function and the conditioning of the training data kernel matrix. We show on real and synthetic data that compared to cross-validation and marginal likelihood maximization, our method is on pair in terms of model performance, but up to six orders of magnitude faster.
Submission history
From: Oskar Allerbo [view email][v1] Tue, 24 May 2022 10:36:05 GMT (1375kb,D)
[v2] Wed, 8 Feb 2023 11:41:24 GMT (2615kb,D)
[v3] Wed, 17 May 2023 12:02:30 GMT (1713kb,D)
[v4] Fri, 1 Dec 2023 13:53:37 GMT (1649kb,D)
Link back to: arXiv, form interface, contact.