Interpolation and Learning with Scale Dependent Kernels

Abstract

We study the learning properties of nonparametric ridge-less least squares. In particular, we consider the common case of estimators defined by scale dependent kernels, and focus on the role of the scale. These estimators interpolate the data and the scale can be shown to control their stability through the condition number. Our analysis shows that are different regimes depending on the interplay between the sample size, its dimensions, and the smoothness of the problem. Indeed, when the sample size is less than exponential in the data dimension, then the scale can be chosen so that the learning error decreases. As the sample size becomes larger, the overall error stop decreasing but interestingly the scale can be chosen in such a way that the variance due to noise remains bounded. Our analysis combines, probabilistic results with a number of analytic techniques from interpolation theory.

arXiv commentThe paper is not completed and contains parts which need to be modified

References in corpus (6)

• Reconciling modern machine learning practice and the bias-variance trade-off
• Benign Overfitting in Linear Regression
• To understand deep learning we need to understand kernel learning
• Does data interpolation contradict statistical optimality?
• On the Sample Complexity of Subspace Learning
• A Closer Look at Covering Number Bounds for Gaussian Kernels

Cited by in corpus (5)

• Kernel regression in high dimensions: Refined analysis beyond double descent
• PSD Representations for Effective Probability Models
• Spectral Analysis of the Neural Tangent Kernel for Deep Residual Networks
• Generalization error of minimum weighted norm and kernel interpolation
• Online nonparametric regression with Sobolev kernels