Id: arxiv/2608.24283 Url: https://arxiv.org/abs/2608.24283 Tags: math.NA Score: 0 Title: Samplet compression for conditionally positive definite kernels and universal Kriging Domain: math.NA Source: arxiv Authors: Sara Avesani, Rüdiger Kempf, Michael Multerer, Holger Wendland Reviews: 0 Subject: math.NA Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-08-25 Page: https://delphi.science/r/arxiv/2608.24283 We present a samplet-based framework for the efficient numerical solution of saddle-point systems arising from conditionally positive definite (CPD) kernel approximation in general and universal Kriging in particular. The vanishing moment property of samplets as well as the particular structure of the associated scaling distributions, which correspond to discrete orthogonal polynomials, allow for a numerically favorable representation of these saddle-point systems. Concretely, they enable a natural null-space reduction of the indefinite saddle-point system to a (large) linear system for the…