The structure of the polytope algebra

From MaRDI portal
Publication:1360367





The polytope algebra \(\Pi\) is the universal group for translation-invariant valuations on convex polytopes in a finite-dimensional real vector space. This group is endowed with a multiplication, via Minkowski sum of polytopes, and the action of the group of dilations on \(\Pi\) can be diagonalized; this almost makes \(\Pi\) a graded algebra over \({\mathbf R}\) [see \textit{P. McMullen}, Adv. Math. 78, 76-130 (1989; Zbl 0686.52005)]. In the present paper, the following structure theorem for \(\Pi\) is obtained. Let \(V\) be a finite-dimensional real vector space, let \(R\) be the algebra of continuous, piecewise polynomial functions on \(V\), and let \(\overline R\) be the quotient of \(R\) by its ideal generated by coordinate functions; then the polytope algebra associated to the dual space \(V^*\) is isomorphic to the image in \(\overline R\) of functions with integral value at 0. This explains the nontrivial grading of \(\Pi\) by the obvious grading of \(R\); separation and duality results for \(\Pi\) can be recovered in this way, too.




Cited in
(44)








This page was built for publication: The structure of the polytope algebra

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1360367)