Diffusion Maps Kernel Ridge Regression

DOI10.48550/arXiv.2608.09623MaRDI QIDQ7284700arXivFDO

Authors John Harlim, Daning Huang, Jiwoo Song

Publication date 10 August 2026

Published in arXiv

Tldr We study kernel ridge regression using diffusion maps kernels constructed from data. We show that these kernels converge to heat kernels on underlying manifolds and obtain corresponding risk bounds, outperforming Gaussian kernels for supervised learning tasks on manifolds.

Full work available at URL https://doi.org/10.48550/arXiv.2608.09623

arXiv classification

math.STmath.NA

Cites work

• Consistency of spectral clustering
• Diffusion maps
• Eigenvalues of the Laplacian on a compact manifold with density
• Ghost Point Diffusion Maps for Solving Elliptic PDEs on Manifolds with Classical Boundary Conditions
• Kernel flows: from learning kernels from data into the abyss
• Learning dynamical systems from noisy data with Weak-form Kernel Ridge Regression
• Learning the kernel matrix with semidefinite programming
• Lipschitz regularity of graph Laplacians on random data clouds
• Mathematical foundations of infinite-dimensional statistical models
• Radial basis approximation of tensor fields on manifolds: From operator estimation to manifold learning
• Rates of strong uniform consistency for multivariate kernel density estimators. (Vitesse de convergence uniforme presque sûre pour des estimateurs à noyaux de densités multivariées)
• Scattered data interpolation on embedded submanifolds with restricted positive definite kernels: Sobolev error estimates
• Spectral convergence of graph Laplacian and heat kernel reconstruction in L^ from random samples
• Spectral convergence of symmetrized graph Laplacian on manifolds with boundary
• Support Vector Machines

This page was built for publication: Diffusion Maps Kernel Ridge Regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7284700)