Approximation of full Lagrange functions by local Lagrange functions

From papers

Let G=(V,E)G=(\mathcal{V},E) be the graph underlying the neighborhood B(v0,K)B(v_0,K), let V~V\widetilde{\mathcal{V}}\subset\mathcal{V} be the interpolation set, and let χ(,v0)\chi(\cdot,v_0) and χˉ(,v0)\bar{\chi}(\cdot,v_0) denote respectively the full and local Lagrange functions centered at v0V~v_0\in\widetilde{\mathcal{V}}, with K>0K>0. Approximation conjecture. The local Lagrange functions can be made arbitrarily close to the full Lagrange functions with a relatively small value of KK. This is intended to justify replacing the computationally expensive full Lagrange functions by local ones, but the statement does not specify a quantitative error bound or a precise meaning of “relatively small”; its resolution is therefore unclear.

Progress summary

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Sources & referencesView supporting material

Primary source

John Paul Ward, Francis J. Narcowich and Joseph D. Ward, “Interpolating splines on graphs for data science applications”, arXiv:1806.10695 (2020).

Solutions 0

No solutions have been posted yet.