ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let x⊂Ω⊂Rn\boldsymbol{x}\subset\Omega\subset\mathbb{R}^n be any set of distinct points, and let A(x,ε)\boldsymbol{A}(\boldsymbol{x},\varepsilon) be the RBF interpolation matrix. For varepsilon1>ε2varepsilon_1>\varepsilon_2, set

A1=A(x,ε1),A2=A(x,ε2).A_1=\boldsymbol{A}(\boldsymbol{x},\varepsilon_1),\qquad A_2=\boldsymbol{A}(\boldsymbol{x},\varepsilon_2).

Condition-number monotonicity conjecture. For the Gaussian or inverse multiquadric RBF,

cond⁡(A1)≤cond⁡(A2).\operatorname{cond}(A_1)\leq\operatorname{cond}(A_2).

The claim formalizes the observed decrease of the matrix condition number as the shape parameter increases. It was proved in the source only for the case N=2N=2; a proof for arbitrary sets of distinct points and dimensions remains open.

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.