The paper introduces a neighboring early‑stopping rule for adaptive regularization in kernel ridge regression with random features (KRR‑RF). By using a uniform grid in inverse regularization and comparing only adjacent estimators, the method reduces discrepancy checks and can be computed directly in the random‑feature space without forming the full kernel Gram matrix. Under standard source and capacity assumptions, the selected estimator achieves the oracle polynomial learning rate up to logarithmic factors, enabling regularization selection without prior knowledge of smoothness or capacity exponents.
arXiv:2609.21017v1 Announce Type: cross Abstract: We study data-driven early stopping for spectral regularisation methods in the classical non-parametric regression setting. Building on the discrepan...
The paper investigates fixed‑design random feature ridge regression beyond the mean‑kernel approximation, focusing on how the predictor’s nonlinear dependence on the empirical kernel affects training error, test error, and the conditional generalization gap. By deriving covariance‑level (one‑loop) corrections via a finite resolvent identity, the authors avoid an almost‑sure Neumann‑series assumption and provide explicit remainder bounds for training error. Numerical experiments demonstrate that including mixed train–test covariance tensors is essential for accurate test error predictions and reveal a width–regularization boundary where second‑order truncation becomes unreliable.
arXiv:2606. 26975v1 Announce Type: cross Abstract: Empirical Bayes (EB) estimators can match the first-order asymptotic risk of maximum likelihood (ML) while behaving very differently at second order: recent excess mean squared error (XMSE) analysis shows that kernel-based EB estimation may be worse than ML when the kernel is poorly aligned with the true parameter.
The paper studies random projections (RP) for spectral algorithms, providing a comprehensive analysis of their generalization performance under broad conditions. By exploiting the RKHS embedding property and operator similarity, it establishes optimal learning rates in Sobolev norms that match minimax lower bounds up to logarithmic factors. For common RP implementations—randomized sketches and Nyström sub-sampling—the required projection dimension scales with the effective dimension, reducing computational cost while preserving statistical efficiency.
arXiv:2303. 08777v3 Announce Type: replace-cross Abstract: Cross-validation is one of the most widely used tools for risk estimation and model selection in statistics and machine learning, yet its theoretical properties when embedded in a learning procedure remain insufficiently understood.
arXiv:2607. 07513v1 Announce Type: new Abstract: Self-supervised learning matches supervised accuracy from a fraction of the labels, but the labeled-sample efficiency behind this has lacked a theoretical explanation.
Empirical Bayes (EB) estimators can match the first-order asymptotic risk of maximum likelihood (ML) while behaving very differently at second order: recent excess mean squared error (XMSE) analysis shows that kernel-based EB estimation may be worse than ML when the kernel is poorly aligned with the true parameter. This paper turns that diagnostic into a design principle.
arXiv:2406. 04425v2 Announce Type: replace Abstract: A fundamental problem in machine learning is understanding the effect of early stopping on the parameters obtained and the generalization capabilities of the model.
arXiv:2610. 23337v1 Announce Type: new Abstract: Pretrained models are often reused across downstream tasks, motivating the question of whether this transfer learning procedure can provide a substantial performance improvement, namely, in terms of the scaling law.
The paper proves spectral convergence of the random feature method (RFM) for multidimensional targets across various function classes, providing high‑probability approximation estimates that hold simultaneously for all admissible error norms. It extends these results to strong‑ and weak‑form RFM discretizations, yielding convergence guarantees for multidimensional second‑order elliptic boundary value and eigenvalue problems. Additionally, it demonstrates super‑exponential singular‑value decay for Fourier features and exponential decay for tanh features, while establishing corresponding condition‑number lower bounds, highlighting a trade‑off between accuracy and ill‑conditioning.
arXiv:2605.15240v2 Announce Type: replace-cross Abstract: This paper investigates the critical role of eigenalignments between the kernel matrix and learning targets in achieving robust generalizatio...
arXiv:2606. 27298v1 Announce Type: cross Abstract: We study the fundamental problem of learning a high-dimensional Gaussian truncated to an unknown halfspace.