Positivity conjecture for principal minors of the bivariate Bézier collocation matrix

Let dd be a nonnegative integer, let TT be a triangle, and let Id\mathcal I_d be the set of weak 33-compositions of dd. For each \bfmiId\bfm i\in\mathcal I_d, let ξ\bfmi\xi_{\bfm i} be the corresponding domain point in barycentric coordinates, and let B\bfmjdB^d_{\bfm j} be the degree-dd Bernstein basis polynomial. For a nonempty subset Γ={\bfmi1,,\bfmin}Id\Gamma=\{\bfm i_1,\dots,\bfm i_n\}\subset\mathcal I_d, define

MΓ=[B\bfmjd(ξ\bfmi)]\bfmi,\bfmjΓ.M_\Gamma=[B^d_{\bfm j}(\xi_{\bfm i})]_{\bfm i,\bfm j\in\Gamma}.

The principal-minor positivity conjecture. For every such Γ\Gamma, the matrix MΓM_\Gamma is nonsingular and

detMΓ>0.\det M_\Gamma>0.

The conjecture asserts unique solvability and positivity for all constrained Lagrange interpolation problems based on these Bernstein collocation matrices. The paper states that the full conjecture is confirmed in its results, including arbitrary ΓId\Gamma\subset\mathcal I_d for d17d\leq17; the supplied status is unknown, so its resolution beyond those cases should be checked.

Sources & referencesView supporting material

Primary source

Gasper Jaklic and Tadej Kanduc, “On positivity of principal minors of bivariate Bezier collocation matrix”, arXiv:1106.0631 (2015).

Progress summary

Never refreshed

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

Solutions 0

No solutions have been posted yet.