MathLabs
TheoremProved

Minkowski's Convex Body Theorem

Statement

Let Λ⊂Rd\Lambda \subset \mathbb{R}^d be a full-rank lattice with fundamental domain volume det⁡(Λ)\det(\Lambda), and let K⊂RdK \subset \mathbb{R}^d be a centrally symmetric convex set (x∈K⇒−x∈Kx \in K \Rightarrow -x \in K). Then vol⁡(K)>2ddet⁡(Λ)  ⟹  K∩(Λ∖{0})≠∅\operatorname{vol}(K) > 2^d \det(\Lambda) \;\Longrightarrow\; K \cap (\Lambda \setminus \{0\}) \neq \emptyset (and if KK is also compact, the strict inequality >> can be weakened to ≥\ge).

Why is it true?

It is the continuous pigeonhole principle of number theory: once a symmetric convex body is larger than 2d2^d fundamental cells of the lattice, shrinking it by a factor of 22 still leaves volume larger than one cell, forcing two points of the shrunk body to differ by a nonzero lattice vector — and symmetry plus convexity pull that lattice vector back inside KK.

Proof sketch

Consider the half-sized body 12K={ x/2:x∈K }\frac{1}{2} K = \{\, x / 2 : x \in K \,\}. Scaling in dd dimensions multiplies volume by 2−d2^{-d}, so the hypothesis vol⁡(K)>2ddet⁡(Λ)\operatorname{vol}(K) > 2^d \det(\Lambda) becomes vol⁡(12K)>det⁡(Λ)\operatorname{vol}(\frac{1}{2} K) > \det(\Lambda).

Let FF be a fundamental parallelepiped of Λ\Lambda, so Rd=⨆v∈Λ(F+v)\mathbb{R}^d = \bigsqcup_{v \in \Lambda} (F + v) and vol⁡(F)=det⁡(Λ)\operatorname{vol}(F) = \det(\Lambda). Cut 12K\frac{1}{2} K into pieces Av=(12K)∩(F+v)A_v = (\frac{1}{2} K) \cap (F + v) and translate each piece back into FF via Bv=Av−v⊆FB_v = A_v - v \subseteq F. Since the sum of the volumes of the BvB_v equals vol⁡(12K)>vol⁡(F)\operatorname{vol}(\frac{1}{2} K) > \operatorname{vol}(F), the sets BvB_v cannot be pairwise disjoint (Blichfeldt's principle).

Pick distinct lattice vectors u≠vu \neq v in Λ\Lambda with Bu∩Bv≠∅B_u \cap B_v \neq \emptyset. Then there exist two distinct points p,q∈12Kp, q \in \frac{1}{2} K such that p−u=q−vp - u = q - v, so p−q=u−v∈Λ∖{0}p - q = u - v \in \Lambda \setminus \{0\}. Because p,q∈12Kp, q \in \frac{1}{2} K, we have 2p,2q∈K2p, 2q \in K; central symmetry of KK gives −2q∈K-2q \in K, and convexity of KK places the midpoint 12(2p)+12(−2q)=p−q\frac{1}{2}(2p) + \frac{1}{2}(-2q) = p - q inside KK. Thus p−qp - q is a nonzero lattice point in K∩(Λ∖{0})K \cap (\Lambda \setminus \{0\}).

Topics that use this theorem

Step-by-step proofs

No step-by-step proof yet for this theorem.

References

  1. Peter M. Gruber (2007). Convex and Discrete Geometry (Grundlehren der mathematischen Wissenschaften, Vol. 336) · DOI:10.1007/978-3-540-71133-9
  2. Maryna S. Viazovska (2017). The sphere packing problem in dimension 8 · arXiv:1603.04246
  3. Henry Cohn, Abhinav Kumar, Stephen D. Miller, Danylo Radchenko, Maryna Viazovska (2019). Universal optimality of the E8 and Leech lattices and interpolation formulas · arXiv:1902.05438