A local maximizer for lattice width of 3-dimensional hollow bodies

Lattice polytopes in convex geometry (including relations with commutative algebra and algebraic geometry) (52B20) Lattices and convex bodies in (n) dimensions (aspects of discrete geometry) (52C07) Integer programming (90C10)

Abstract: The second and fourth authors have conjectured that a certain hollow tetrahedron

D e l t a

of width

2 + s q r t 2

attains the maximum lattice width among all three-dimensional convex bodies. We here prove a local version of this conjecture: there is a neighborhood

U

of

D e l t a

in the Hausdorff distance such that every convex body in

U s e t m i n u s D e l t a

has width strictly smaller than

D e l t a

. When the search space is restricted to tetrahedra, we compute an explicit such neighborhood. We also limit the space of possible counterexamples to the conjecture. We show, for example, that their width must be smaller than

3.972

and their volume must lie in

[2.653, 19.919]

.

Review text

Consider a convex body \(K\) contained in the \(d\)-dimensional Euclidean space \(\mathbb{R}^d\) and a non-zero vector \(v\) in that space. The width of \(K\) along \(v\) is equal to \(\|v\|\) times the length of the orthogonal projection of \(K\) on a straight line spanned by \(v\). Then if \(\Lambda\) is a \(d\)-dimensional lattice contained in \(\mathbb{R}^d\) (for instance, \(\mathbb{Z}^d\)), the lattice width of \(K\) is the least possible width of \(K\) along the non-zero vectors in the dual lattice of \(\Lambda\). A convex body is hollow when its interior is disjoint from \(\Lambda\). The flatness theorem tells that the lattice width of any hollow convex body is bounded by a finite number that only depends on \(d\). This theorem and the largest possible lattice width \(f(d)\) of \(d\)-dimensional hollow convex bodies play an important role in integer programming as reported, for instance by \textit{R. Kannan} and \textit{L. Lovász} [Ann. Math. (2) 128, No. 3, 577--602 (1988; Zbl 0659.52004)]. It is known that \(f(1)=1\) and \textit{C. A. J. Hurkens} has shown that \(f(2)=1+2/\sqrt{3}\) [Linear Algebra Appl. 134, 121--128 (1990; Zbl 0708.52002)]. When \(d=3\), Giulia Codenotti and Francisco Santos have found a hollow tetrahedron \(\Delta\) whose lattice width is \(2+\sqrt{2}\) and they conjecture that \(f(3)\) is equal to this value [\textit{G. Codenotti} and \textit{F. Santos}, Proc. Am. Math. Soc. 148, No. 2, 835--850 (2020; Zbl 1445.52012)]. Here, it is shown that there exists an open set \(U\) in \(\mathbb{R}^3\) that contains \(\Delta\) and such that every hollow convex body contained in \(U\) has lattice width less than \(2+\sqrt{2}\). In other words, \(\Delta\) is a local minimum with respect to lattice width. The proof consists in parametrizing the space of the hollow tetrahedra obtained by perturbing \(\Delta\) and by using Karush-Kuhn-Tucker conditions to search for a local maximum in that space. Necessary conditions are also given for a \(3\)-dimensional hollow convex body to have lattice width greater than \(2+\sqrt{2}\).

Recommendations

Cites work

Cited in

(8)

Describes a project that uses

Uses Software

SageMath

This page was built for publication: A local maximizer for lattice width of 3-dimensional hollow bodies

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2028093)