English

Just Interpolate: Kernel "Ridgeless" Regression Can Generalize

Statistics Theory 2020-07-27 v2 Machine Learning Machine Learning Statistics Theory

Abstract

In the absence of explicit regularization, Kernel "Ridgeless" Regression with nonlinear kernels has the potential to fit the training data perfectly. It has been observed empirically, however, that such interpolated solutions can still generalize well on test data. We isolate a phenomenon of implicit regularization for minimum-norm interpolated solutions which is due to a combination of high dimensionality of the input data, curvature of the kernel function, and favorable geometric properties of the data such as an eigenvalue decay of the empirical covariance and kernel matrices. In addition to deriving a data-dependent upper bound on the out-of-sample error, we present experimental evidence suggesting that the phenomenon occurs in the MNIST dataset.

Keywords

Cite

@article{arxiv.1808.00387,
  title  = {Just Interpolate: Kernel "Ridgeless" Regression Can Generalize},
  author = {Tengyuan Liang and Alexander Rakhlin},
  journal= {arXiv preprint arXiv:1808.00387},
  year   = {2020}
}

Comments

28 pages, 8 figures

R2 v1 2026-06-23T03:21:44.814Z