Prove2Me
Navigate
DiscoverFormalpediaBlogsUsersMomentumMy Missions+
Prove2Me
⌕
Log in
← Formalpedia

The least uniform complex coordinate Hlawka constant for p ≥ 85

Proved
HlawkaSchatten.DiagonalCutoff.cutoff85

by savarin · Oct 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

complex-analysisdiagonal-constructionhlawka-schattenoptimal-constant

For every real exponent p ≥ 85, let N_p(x) be the foundation's finite coordinate p-norm and let K_p be its unchanged cyclicConstant p, defined as the supremum of the cyclic ratio over t ∈ [1/2, 2].

Then K_p is the least real constant C such that the triple deficit N_p(x) + N_p(y) + N_p(z) − N_p(x + y + z) is at most C times the pair-deficit sum 2(N_p(x) + N_p(y) + N_p(z)) − N_p(x + y) − N_p(x + z) − N_p(y + z), for every natural-number dimension n and all complex vectors x, y, z in that dimension.

This combines admissibility and optimality uniformly over all finite complex coordinate dimensions. It includes dimension zero and arbitrary triples with unequal or zero norms. The same shared definitions are used throughout; the claim concerns complex diagonal matrices via their coordinate norms.

No proof is known for 85 ≤ p < 87; the accepted cutoff-87 theorem covers every p ≥ 87.

Preamble
import Definitions.Def_HlawkaSchatten_DiagonalConstruction_Basic
import Definitions.Def_HlawkaSchatten_DiagonalConstruction_Cyclic
import Definitions.Def_HlawkaSchatten_GapComparison
import Mathlib.Analysis.Complex.Circle
import Mathlib.Analysis.Complex.ExponentialBounds
import Mathlib.Analysis.Convex.Deriv
import Mathlib.Analysis.Convex.Function
import Mathlib.Analysis.Convex.Integral
import Mathlib.Analysis.Convex.Jensen
import Mathlib.Analysis.Convex.SpecificFunctions.Basic
import Mathlib.Analysis.Convex.SpecificFunctions.Pow
import Mathlib.Analysis.InnerProductSpace.Basic
import Mathlib.Analysis.InnerProductSpace.Dual
import Mathlib.Analysis.InnerProductSpace.NormPow
import Mathlib.Analysis.Normed.Lp.PiLp
import Mathlib.Analysis.Normed.Module.FiniteDimension
import Mathlib.Analysis.SpecialFunctions.Pow.Continuity
import Mathlib.Data.Fin.VecNotation
import Mathlib.Data.Real.Basic
import Mathlib.Data.Sign.Basic
import Mathlib.LinearAlgebra.Dimension.Finite
import Mathlib.MeasureTheory.Group.Integral
import Mathlib.MeasureTheory.Integral.Bochner.ContinuousLinearMap
import Mathlib.MeasureTheory.Measure.Haar.Basic
import Mathlib.Tactic.Abel
import Mathlib.Tactic.FieldSimp
import Mathlib.Tactic.Linarith
import Mathlib.Tactic.LinearCombination
import Mathlib.Tactic.Module
import Mathlib.Tactic.Positivity
import Mathlib.Tactic.Ring
import Mathlib.Topology.Instances.Sign
import Mathlib.Topology.Order.Compact

/-! # The least dimension-independent complex coordinate Hlawka constant

This packages admissibility and the three-coordinate cyclic obstruction into
one statement, with an explicit lower cutoff on the real exponent.
-/

open HlawkaSchatten
open HlawkaSchatten.DiagonalConstruction
Formal statement
theorem HlawkaSchatten.DiagonalCutoff.cutoff85 :
    ∀ p : ℝ, 85 ≤ p →
      IsLeast {C : ℝ | ∀ n : ℕ,
        HasHlawkaConstant (lpNorm p : (Fin n → ℂ) → ℝ) C}
        (cyclicConstant p) := by sorry
Source
https://prove2.me/campaigns/sharp-diagonal-hlawka-constant
Read-back

What the Lean code literally says, in plain math · claude-opus-5-5

For every real number p≥85p \ge 85p≥85 (ppp need not be an integer, and p=85p = 85p=85 itself is included), the statement asserts that the real number c(p)c(p)c(p) defined below is the least element of the set SpS_pSp​. This set consists of all real numbers CCC such that, for every n∈N={0,1,2,… }n \in \mathbb{N} = \lbrace 0, 1, 2, \dots \rbracen∈N={0,1,2,…} and all vectors x,y,z∈Cnx, y, z \in \mathbb{C}^nx,y,z∈Cn,

Δ3(x,y,z)≤C[Δ2(x,y)+Δ2(x,z)+Δ2(y,z)],\Delta_3(x,y,z) \le C\big[\Delta_2(x,y) + \Delta_2(x,z) + \Delta_2(y,z)\big],Δ3​(x,y,z)≤C[Δ2​(x,y)+Δ2​(x,z)+Δ2​(y,z)],

where

Δ2(u,v)=∥u∥p+∥v∥p−∥u+v∥p,Δ3(x,y,z)=∥x∥p+∥y∥p+∥z∥p−∥x+y+z∥p.\Delta_2(u,v) = \lVert u\rVert_p + \lVert v\rVert_p - \lVert u+v\rVert_p, \qquad \Delta_3(x,y,z) = \lVert x\rVert_p + \lVert y\rVert_p + \lVert z\rVert_p - \lVert x+y+z\rVert_p .Δ2​(u,v)=∥u∥p​+∥v∥p​−∥u+v∥p​,Δ3​(x,y,z)=∥x∥p​+∥y∥p​+∥z∥p​−∥x+y+z∥p​.

Vectors are added coordinatewise. For x=(x1,…,xn)x = (x_1, \dots, x_n)x=(x1​,…,xn​),

∥x∥p=(∑i=1n∣xi∣p)1/p,\lVert x\rVert_p = \Big(\sum_{i=1}^{n} |x_i|^p\Big)^{1/p},∥x∥p​=(i=1∑n​∣xi​∣p)1/p,

where ∣xi∣|x_i|∣xi​∣ is the complex modulus and the powers are real powers. Because p≥85>0p \ge 85 > 0p≥85>0, we have 0p=00^p = 00p=0, so this is the ordinary ℓp\ell^pℓp-norm on Cn\mathbb{C}^nCn. For n=1n = 1n=1 it equals ∣x1∣|x_1|∣x1​∣, and on C0\mathbb{C}^0C0 (a single point) it equals 000. Rearranged, the inequality is equivalent to

C(∥x+y∥p+∥x+z∥p+∥y+z∥p)≤(2C−1)(∥x∥p+∥y∥p+∥z∥p)+∥x+y+z∥p.C\big(\lVert x+y\rVert_p + \lVert x+z\rVert_p + \lVert y+z\rVert_p\big) \le (2C-1)\big(\lVert x\rVert_p + \lVert y\rVert_p + \lVert z\rVert_p\big) + \lVert x+y+z\rVert_p .C(∥x+y∥p​+∥x+z∥p​+∥y+z∥p​)≤(2C−1)(∥x∥p​+∥y∥p​+∥z∥p​)+∥x+y+z∥p​.

Being the least element amounts to two claims:

  1. Membership: the inequality holds with C=c(p)C = c(p)C=c(p) for every nnn and all x,y,z∈Cnx, y, z \in \mathbb{C}^nx,y,z∈Cn.
  2. Minimality: every real CCC for which the inequality holds for every nnn and all x,y,z∈Cnx, y, z \in \mathbb{C}^nx,y,z∈Cn satisfies c(p)≤Cc(p) \le Cc(p)≤C. Equivalently, for each real C<c(p)C < c(p)C<c(p) there exist some nnn (which may depend on CCC) and some x,y,z∈Cnx, y, z \in \mathbb{C}^nx,y,z∈Cn for which the inequality fails. This claim concerns constants that are valid in all dimensions at once. Nothing is asserted about the best constant in any single fixed dimension.

The constant. For real ttt, define

Ap(t)=(tp+2)1/p,Bp(t)=(2∣1−t∣p+2p)1/p,Rp(t)=3Ap(t)−31/p∣2−t∣6Ap(t)−3Bp(t),A_p(t) = (t^p + 2)^{1/p}, \qquad B_p(t) = \big(2|1-t|^p + 2^p\big)^{1/p}, \qquad R_p(t) = \frac{3A_p(t) - 3^{1/p}|2-t|}{6A_p(t) - 3B_p(t)},Ap​(t)=(tp+2)1/p,Bp​(t)=(2∣1−t∣p+2p)1/p,Rp​(t)=6Ap​(t)−3Bp​(t)3Ap​(t)−31/p∣2−t∣​, c(p)=sup⁡{Rp(t):12≤t≤2}.c(p) = \sup \lbrace R_p(t) : \tfrac12 \le t \le 2 \rbrace .c(p)=sup{Rp​(t):21​≤t≤2}.

The supremum is taken over the closed interval, both endpoints included. On that interval t>0t > 0t>0 and ∣2−t∣=2−t|2-t| = 2-t∣2−t∣=2−t, and c(p)c(p)c(p) depends only on ppp.

In the formal system, a quotient with zero denominator equals 000, and the supremum of an empty or unbounded-above set of reals also equals 000. Neither convention applies here. For 12≤t≤2\tfrac12 \le t \le 221​≤t≤2 we have ∣1−t∣≤1|1-t| \le 1∣1−t∣≤1, so

Bp(t)p=2∣1−t∣p+2p≤2+2p<2p(tp+2)=(2Ap(t))p.B_p(t)^p = 2|1-t|^p + 2^p \le 2 + 2^p < 2^p(t^p+2) = (2A_p(t))^p .Bp​(t)p=2∣1−t∣p+2p≤2+2p<2p(tp+2)=(2Ap​(t))p.

Hence the denominator 3(2Ap(t)−Bp(t))3\big(2A_p(t) - B_p(t)\big)3(2Ap​(t)−Bp​(t)) is strictly positive and RpR_pRp​ is continuous on the interval. So c(p)=max⁡1/2≤t≤2Rp(t)c(p) = \max_{1/2 \le t \le 2} R_p(t)c(p)=max1/2≤t≤2​Rp​(t), and this maximum is attained.

Scope and edge cases.

  • The only hypothesis is p≥85p \ge 85p≥85. There are no other parameters or assumptions, and nothing is asserted for p<85p < 85p<85.
  • n=0n = 0n=0 is included. There every norm is 000, so the inequality reads 0≤00 \le 00≤0, which holds for every CCC and imposes no constraint.
  • One and the same CCC must work for every nnn simultaneously.
  • x,y,zx, y, zx,y,z are arbitrary complex vectors, including zero, equal or parallel vectors.
  • In both claims, CCC ranges over all real numbers with no sign restriction.
Human review
  • Endorsed by marwahaha · Oct 5, 2026

  • Endorsed by savarin · Oct 5, 2026

    Confirmed by the mission captain (proposal self-audit).

  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paper
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me