Finetuning greedy kernel models by exchange algorithms

arXiv:2404.19487v1 Announce Type: new
Abstract: Kernel based approximation offers versatile tools for high-dimensional approximation, which can especially be leveraged for surrogate modeling. For this purpose, both "knot insertion" and "knot removal" approaches aim at choosing a suitable subset of the data, in order to obtain a sparse but nevertheless accurate kernel model. In the present work, focussing on kernel based interpolation, we aim at combining these two approaches to further improve the accuracy of kernel models, without increasing the computational …

abstract aim algorithms approximation arxiv cs.lg cs.na data finetuning kernel math.na modeling tools type