{"id":"arxiv/2611.09612","url":"https://arxiv.org/abs/2611.09612","tags":["stat.ML","cs.LG","math.ST"],"score":0,"title":"Kernel-Based Learning of Analytic Operators and Parametric PDEs","domain":"stat.ML","source":"arxiv","authors":["Giovanni S. Alberti","Markus Holzleitner","Lorenzo Rosasco","Matteo Santacesaria"],"reviews":0,"subject":"stat.ML","summary":"We study kernel-based learning of analytic operators from finite samples, with a focus on how the regularity of the target should determine the choice of the kernel. We establish embeddings of analytic operators into vector-valued reproducing kernel Hilbert spaces and derive finite-sample, high-probability guarantees that quantify the uncertainty due to random sampling: in the noiseless setting, suitable analytic kernels yield rates exponential in a fractional power of the sample size, while Sobolev-type kernels yield polynomial rates. We apply the theory to the solution operator of a…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-11-06","reactions":{}}