Proposition 1
DD7VIf D is symmetric about the origin and convex, then 2D is convex and symmetric about the origin.
Extremal functions on moduli spaces and applications
- arXiv
- 2411.13671v3 · PDF · 23 December 2024
- Categories
- math.NT
- Licence
- CC BY SA 4.0 The site shows the text.
Abstract
Our object of study is extremal functions which are defined by distance functions of convex bodies. These functions take values in the moduli spaces of algebraic and geometric objects associated with these {\mathbb Z}-modules (geometric lattices) and with convex bodies. In most cases, convex bodies are 2-dimensional Minkowski balls whose boundaries are Minkowski curves and we study lattice points on these curves. We define and investigate extremal functions that yield the homogeneous arithmetic minimum of a function in a lattice, the Hermite constant, the critical determinant of a body, optimal packings of bodies, best values of covering constants, and optimal solutions of Diophantine approximation problems. Moreover, for two-dimensional unit Minkowski balls and Minkowski domains we determine the minimal areas of inscribed and circumscribed hexagons.
Items 24
Corollary 1
O9UPLet m be integer m \ge 0 and n be natural greater m. If 2^m D centrally symmetric convex body then 2^n D is again centrally symmetrc convex body.
Definition 1
DUBUWe will call the one-dimensional Minkowski sphere (?) the Minkowski curve.
For natural p, d, p = 2d the Minkowski curves
{\mathbb S}^{1}_{2d} = C_{2d}: x^{2d} + y^{2d} = 1
are algebraic curves. We will call these curves Minkowski algebraic curves (algebraic curves, if it is clear from the context which curves are being considered).
Remark 1
FXOIOver complex numbers {\mathbb C} the projectivization of curve C_{2d} with polynomial equation p(x,y) = x^{2d} + y^{2d} - 1 \in {\mathbb C}[x,y] defines the projective curve of the form x^{2d} + y^{2d} - z^{2d} = 0 in the projective space {\mathbb C}{\mathbb P}^2. Its genus g = (2d - 1)(d - 1).
Proposition 2
ZWVHFrom theorem
[GGM:PM] (see also [gl1]) we have next expressions for critical determinants and their lattices:
{\Delta^{(0)}_p} = \Delta(p, {\sigma_p}) = \frac{1}{2}{\sigma}_{p},
{\sigma}_{p} = (2^p - 1)^{1/p},
{\Delta^{(1)}_p} = \Delta(p,1) = 4^{-\frac{1}{p}}\frac{1 +\tau_p }{1 - \tau_p},
2(1 - \tau_p)^p = 1 + \tau_p^p, \; 0 \le \tau_p < 1.
Used by Remark 8
Example 1
QV0ALattice \Lambda_{p}^{(0)} are two-dimensional lattices in {\mathbb R}^2 spanned by the vectors
\lambda^{(1)} = (1, 0),
\lambda^{(2)} = (\frac{1}{2}, \frac{1}{2} \sigma_p).
For lattices \Lambda_{p}^{(1)}, due to the cumbersomeness of the formulas for the coordinates of basis vectors of the lattices, as an example, we present only the lattice \Lambda_{2}^{(1)}. The lattice \Lambda_{2}^{(1)} is a two-dimensional lattice in {\mathbb R}^2 spanned by the vectors
\lambda^{(1)} = (-2^{-1/2},2^{-1/2}),
\lambda^{(2)} = (\frac{\sqrt 6 - \sqrt 2}{4}, \frac{\sqrt 6 + \sqrt 2}{4}).
Lemma 1
HQFWLet (P_x, P_y) be a point of the critical lattice \Lambda on Minkowski curve {\mathbb S}^{1}_{p} = C_{p}.
Then the point (u, v) that satisfies conditions
\left\{
\begin{array}{lc}
|P_x v - P_y u| = \; d(\Lambda),\\
|u|^p + |v|^p = \; 1.\\
\end{array}
\right.
belongs to \Lambda and lies on C_{p}. The shell of points of the critical lattice \Lambda_{p} on Minkowski curve C_{p} contains 6 points Point coordinates (u, v) can be calculated in closed form or with any precision.
Example 2
7A0LEach shell of points of the critical lattices \Lambda_{p}^{(0)} on Minkowski curves C_{p} contain 6 points:
\pm (1, 0),
\left(\pm \frac{1}{2}, \pm \frac{1}{2} \sigma_p \right).
Respectively the shell of points of the critical lattice \Lambda_{2}^{(1)} on Minkowski curves C_{p} contains 6 points:
(-2^{-1/2},2^{-1/2}), (2^{-1/2},-2^{-1/2}),
\pm \left(\frac{\sqrt 6 + \sqrt 2}{4}, \frac{\sqrt 6 - \sqrt 2}{4}\right),
\pm \left(\frac{\sqrt 6 - \sqrt 2}{4}, \frac{\sqrt 6 + \sqrt 2}{4} \right).
Remark 2
WT4NThe only integer quadratic forms of the lattices \Lambda_{p}^{(0)} and \Lambda_{p}^{(1)} are the quadratic forms corresponding to the lattices \Lambda_{2}^{(0)} and \Lambda_{2}^{(1)}. These forms are the form
x^2 - xy +y^2
for the lattice \Lambda_{2}^{(0)} with the base \lambda^{(1)} = (1, 0), \lambda^{(2)} = (-\frac{1}{2}, \frac{1}{2} \sqrt 3) and the form
x^2 -xy +y^2
for the lattice \Lambda_{2}^{(1)} with the base \lambda^{(1)} = (-2^{-1/2},2^{-1/2}), \lambda^{(2)} = (\frac{\sqrt 6 + \sqrt 2}{4}, \frac{\sqrt 6 - \sqrt 2}{4}) .
Remark 3
9HN0The length \ell of the Minkowski curve for p > 1 is
{\ell} = 4 \int_0^1 (1 - y^p)^{\frac{1}{p}} dy
Remark 4
DHWSAs \ell > 3 by results of the Jarnic [jar] the curves C_p and for integer N > 0 their N-fold magnifications N\cdot C_p about the origin for corresponding length \ell contains at most
3(4\pi)^{-\frac{1}{3}} {\ell}^{\frac{2}{3}} + {\mathcal O}({\ell}^{\frac{1}{3}})
integral lattice points.
Remark 5
YLJEThe curve C_p for real p > 1 is three times continuity differentiable. From results of Swinnerton–Dyer [sdy] follow that the number of rational points (\frac{m}{N},\frac{n}{N}) on C_p (the number of integral points on N\cdot C_p) as N \to \infty is estimatrd by
|N \cdot C_p \cap {\mathbb Z}^2| \le c(C_p, \epsilon)N^{\frac{3}{5} + \epsilon}.
Let y = f(x) be the arc A of C_{p} and f(x) be a transcendental analytic function on \left[0, 1\right]. Then
|tA \cap {\mathbb Z}^2| \le c(f,\epsilon) t^{\epsilon}.
It follows from Bombieri-Pila formula[bop].
Remark 6
99QVFor algebraic curve C_{2d}({\mathbb C}) its Riemann surface is oriented sphire with g handles. The function field {\mathbb F} on the Riemann surface is the field of rational functions of the genus g. If two such fields are isomorphic then the corresponding algebraic curves are birationaly isomorphic.
Remark 7
0GSJLet (C_{2d}, {\mathcal L}) be a projective variety over K with ample line bundle {\mathcal L} over C_{2d} and let h_{\mathcal L}(\cdot) be a height function assosiated to it. Let \mathcal A be a non compact Riemann surface. For relatively compact open set U \subset {\mathcal A} and for holomorphic map \varphi: \mathcal A \to C_{2 d}({\mathbb C}) with Zariski dense set \varphi(\mathcal A) and the set S_U(T) = \{ z \in U | \varphi(z) \in C_{2 d}(K)\} and h_{\mathcal L}(\varphi(z)) \le T for cardinality \#S_U(T)of the set S_U(T) and for any positive number \epsilon we have
\#S_U(T) \ll \exp (\epsilon T)
where the involved constants depend on U, \varphi, and {\mathcal L} but not on T .
Proposition 3
XQGIThe Minkowski-Cohn moduli space for admissible lattices of
doubled Minkowski balls 2 D_p (Minkowski domains 2 D_p, \; m = 1) has the form
\Delta(p,\sigma)_{2 D_p} = 4 (\tau + \sigma)(1 + \tau^{p})^{-\frac{1}{p}}
(1 + \sigma^p)^{-\frac{1}{p}},
in the same domain (?).
Proposition 4
N3PINow put m \in {\mathbb N}. The moduli spaces of determinants of packing lattices of Minkowski balls and Minkowski domains have the form
\Delta(p,\sigma)_{2^m D_p} = 4^m (\tau + \sigma)(1 + \tau^{p})^{-\frac{1}{p}}
(1 + \sigma^p)^{-\frac{1}{p}}.
Proposition 5
O7TCThe set of areas of the entire family of al-hexagons {\mathcal{H}}_p, (1 < p < \infty) inscribed in Minkowski balls D_p is parameterized by the function A(\sigma, p) over the domain (?) and defining the corresponding moduli space \mathbf{A}:
A(\sigma, p) = 3(\tau + \sigma)(1 + \tau^{p})^{-\frac{1}{p}}
(1 + \sigma^p)^{-\frac{1}{p}},
Used by Remark 8
Corollary 2
JC5WThe moduli space of areas of al-hexagons inscribed in unit Minkowski balls has the form (5).
Proposition 6
NQ7AThe set of areas of the entire family of ( {\mathcal{H}}_p, (1 < p < \infty) circumscribed to unit Minkowski balls has the form:
{\mathbf {AC}}(p,\sigma)_{2 D_p} = 4 (\tau + \sigma)(1 + \tau^{p})^{-\frac{1}{p}}
(1 + \sigma^p)^{-\frac{1}{p}},
in the same domain (?).
Theorem 1
UPFJLet m \in {\mathbb N} (see [gl1] for the case m=1). The critical determinants of 2^m dubling
balls D_p have a representation of the form
{\Delta^{(0)}_p}(2^m D_p) = \Delta(p, {\sigma_p})_{2^m D_p} = 2^{2 m-1}\cdot {\sigma}_{p}, \; {\sigma}_{p} = (2^p - 1)^{1/p},
{\Delta^{(1)}_p}(2^m D_p) = \Delta(p,1)_{2^m D_p} = 4^{m - \frac{1}{p}}\frac{1 +\tau_p }{1 - \tau_p}, \; 2(1 - \tau_p)^p = 1 + \tau_p^p, \; 0 \le \tau_p < 1.
And these are the determinants of the sublattices of index 2^m of the critical lattices of the corresponding balls D_p.
The extremal functions for packing of Minkowski balls and domains have the forms:
\Delta(2^m D_p) = \left\{
\begin{array}{lc}
\Delta^{(1)}_p (2^m D_p), \; 1 < p \le 2, \; p \ge p_{0},\\
\Delta^{(0)}_p (2^m D_p), \; 2 \le p \le p_{0};\\
\end{array}
\right.
Problem
AGOMWhat are extremal function and covering constants for covering problem?
Theorem 2
XDWH\min({ A}({\mathcal {IH}}_p)(D_p)) = \left\{
\begin{array}{lc}
3\cdot4^{-1/p} \frac{1 +\tau_p }{1 - \tau_p}, \; 1 < p \le 2, \; p \ge p_{0},\\
\frac{3}{2}{\sigma}_{p}, \; 2 \le p \le p_{0};\\
\end{array}
\right.
Remark 8
6OGEFrom Propositions 2 and 5 it follows that the moduli spaces of packing lattices and hexagons circumscribed around Minkowski balls coincide. Consequently, their extremal functions also coincide. This function is a specialization of function (1) for m = 1. For completeness, below we present its form.
Theorem 3
I3KZ
Below 0 \le \tau_p < 1,
\min({ A}({\mathcal {CH}}_p)(D_p)) = \left\{
\begin{array}{lc}
4^{1 - \frac{1}{p}}\frac{1 +\tau_p }{1 - \tau_p}, \; 2(1 - \tau_p)^p = 1 + \tau_p^p, \; 1 < p \le 2, \; p \ge p_{0},\\
2 \cdot {\sigma}_{p}, \; {\sigma}_{p} = (2^p - 1)^{1/p},, \; 2 \le p \le p_{0};\\
\end{array}
\right.
Source and licence
The text on this page is from arXiv:2411.13671v3 by Nikolaj Glazunov, under the licence CC BY SA 4.0. The corpus made these changes: it read the items out of the LaTeX source, it expanded the macros of the paper, it put each proof after its statement, and it converted the text to HTML. The formulas are the TeX of the paper, and KaTeX renders them. A formula or a reference can look different from the PDF. The page is published under CC BY SA 4.0.
The record of each item is in analysis/objects/2411/2411.13671.jsonl.