Source-linked AI summary

A Finite Certificate for the Positive $n=11$ Vasc Inequality

Tiancheng Xu, Wensheng Yu

arXiv:2610.04549v1cs.SC

TL;DR

The Vasc cyclic expression depends on variable order, so the inequality must be established across the normalized rank-word cases. The proof rotates the minimum to the first position, represents ordered values by rank words and cumulative gaps, and verifies polynomial certificates on the resulting domains. The verified root family is partitioned into accepted classes, and the proof establishes the inequality for distinct positive tuples.

  • Problem

    The Vasc cyclic expression depends on variable order, so the inequality must be established across the normalized rank-word cases.

    Source: §1.1 · p. 1, §3.4 · p. 8
  • Method

    The proof rotates the minimum to the first position, represents ordered values by rank words and cumulative gaps, and verifies polynomial certificates on the resulting domains.

    Source: §1.4 · p. 2, §3.4 · p. 8
  • Results

    The verified root family is partitioned into accepted classes, and the proof establishes the inequality for distinct positive tuples.

    Source: §3.3.1 · p. 6, §3.4 · p. 8
  • Takeaways & Limitations

    Theorem 1 is stated for every strictly positive eleven-tuple, with subscripts taken modulo eleven.

    Source: §1.3 · p. 2

    The rank-word representation uses strictly positive gaps for distinct-coordinate tuples; nonnegative gaps extend the polynomial certificates to ties and zero values.

    Source: §2.2 · p. 3

Abstract

from arXiv · show

We establish the Vasc cyclic inequality for all strictly positive real eleven-tuples by a finite exact certificate. Cyclic rotation places a minimum at the first position, and rank words together with cumulative gaps reduce the problem to $10!=3{,}628{,}800$ homogeneous integer polynomials of degree eleven. Independent coefficient reconstruction proves $3{,}358{,}617$ roots directly. The remaining $270{,}183$ roots are covered by $267{,}952$ ordinary midpoint certificates, $2{,}195$ complete binary subdivision certificates, and $36$ complete certificates containing partial sorting nodes. Each nontrivial leaf is checked by subtracting weighted AM-GM midpoint circuits from a positive multiple of the transformed polynomial and verifying every residual coefficient. Explicit word-set comparisons certify disjointness and exact coverage. We prove the soundness of the leaves and both subdivision rules, then recover the rational inequality by positive denominator clearing and continuity. The resulting proof combines explicit mathematical reductions with independently checked exact integer certificates.

1 Setup

The paper states the Vasc cyclic inequality for positive eleven-tuples and gives a finite-certificate proof strategy based on rank words and cumulative gaps. It situates this approach among prior inequality certificates and reports coverage of all normalized rank words.

1.1 Convention

  • The Vasc expression is order-dependent, but is invariant under cyclic rotation and common positive scaling.Source: §1.1 · p. 1

1.2 Related work

  • Prior certificate approaches include successive difference substitutions, structured column-stochastic substitutions, Pólya-related methods, and AM–GM-based certificates.Source: §1.2 · pp. 1–2
  • Earlier work covers the positive n=9 case, while a reported n=11 result concerns only a small cone; this paper partitions all normalized rank words.Source: §1.2 · p. 2

1.3 Statement

  • Theorem 1 states the Vasc cyclic inequality for every strictly positive eleven-tuple, and every constant positive tuple attains equality.Source: §1.3 · p. 2

1.4 Proof strategy

  • The proof reduces the inequality to homogeneous polynomials indexed by normalized rank words, then uses nonnegative coefficients or midpoint decompositions on subdivided gap cones.A finite certificate assembles these cones before recovering the rational inequality.Source: §1.4 · p. 2

2 Reductions

Positive denominator clearing converts the rational inequality into an equivalent homogeneous integer-polynomial nonnegativity problem. For distinct-coordinate tuples, cyclic rotation and rank words organize the cases, while cumulative gaps encode their ordered values and also support certificates on boundary cases.

2.1 Clearing the positive denominators

  • Multiplying by the positive denominator product yields an integer polynomial whose sign is equivalent to that of the rational expression.The polynomial is homogeneous of degree eleven.Source: §2.1 · p. 3

2.2 Normalized rank words

  • After rotating a distinct-coordinate tuple so its minimum is first, its rank word records the ranks at the original cyclic positions; there are 10! such words.Source: §2.2 · p. 3
  • Cumulative gaps represent weakly increasing positive lists, with nonnegative gaps allowing ties and zero values in the polynomial certificates.The gap-to-list construction and its inverse are given by adjacent differences and telescoping sums.Source: §2.2 · p. 3
  • For each rank word, substituting its linear forms defines a polynomial on the strictly positive gap domain.The passage also notes that some possible degree-eleven monomial coefficients are zero.Source: §2.2 · p. 4

3 Polynomial nonnegativity certificates

The certificate evaluates each leaf after a path-dependent nonnegative integer transformation of the gap variables. The stated monomial convention also fixes how zero exponents are evaluated.

  • Each leaf certifies its defining polynomial on a transformed orthant, using g = Tz for a path-reconstructed nonnegative integer matrix T.Source: §3 · p. 4
  • The transformed variables z lie in R11, and the leaf certificate concerns nonnegative values on the transformed orthant.Source: §3 · p. 4
  • Monomials use the convention 0^0 = 1.Source: §3 · p. 4

3.1 Leaf certificates

Weighted midpoint circuits provide nonnegative polynomial summands through an exponent identity and an integer inequality on their weights. The exact leaf test either verifies coefficientwise nonnegativity directly or checks a residual after subtracting circuits from a positive multiple of the transformed polynomial.

3.1.1 Weighted midpoint circuits

  • A weighted midpoint circuit uses positive integer weights and exponent vectors satisfying the stated midpoint relation and degree condition.In a degree-(11 + k) leaf, all three exponents have degree 11 + k.Source: §3.1.1 · p. 4
  • Every circuit satisfying the midpoint relation and d^2 ≤ 4uv is nonnegative on R11.Source: §3.1.1 · p. 4

3.1.2 The exact leaf test

  • A leaf is accepted either when every transformed-polynomial coefficient is nonnegative or when every residual coefficient is nonnegative after circuit subtraction.The residual is formed for k in {0, 1, 2, 3}; its coefficients are obtained by exact expansion.Source: §3.1.2 · p. 5
  • An accepted leaf proves the transformed polynomial is nonnegative throughout the nonnegative orthant.Source: §3.1.2 · p. 5

3.2 Finite covers of gap cones

Finite trees cover the gap orthant through binary comparisons or partial sorting, with accepted leaves establishing nonnegativity throughout the covered domain.

3.2 Finite covers of gap cones

  • A cover begins at the identity matrix, and every internal node must have the prescribed children and complete accepted subtrees.Child images may overlap on their boundaries.Source: §3.2 · p. 5, §3.2.2 · p. 6

3.2.1 Binary comparison nodes

  • Binary comparison splits on either ordering of two coordinates, using substitutions that produce two children covering the parent orthant.The child matrices are computed from the parent matrix by adding one selected column to another.Source: §3.2.1 · p. 5

3.2.2 Partial sorting nodes

  • Partial sorting orders a selected group of 2–6 coordinates and creates one child for each permutation, covering the parent orthant.The substitutions use successive coordinate differences; ties can place a point in multiple children.Source: §3.2.2 · p. 6
  • If every leaf is accepted, induction over the finite tree proves the root polynomial nonnegative for every nonnegative gap vector.The induction combines leaf soundness with the coverage of each binary or partial sorting node.Source: §3.2.2 · p. 6

3.3 The complete finite certificate

The certificate partitions the full rank-word family into four accepted classes, verifies the polynomial and leaf conditions by exact computation, and records arithmetic bounds and workload.

3.3.1 The rank-word partition

  • Every rank word belongs to one of four disjoint classes: coefficientwise-nonnegative roots, ordinary midpoint leaves, binary trees, or trees with partial sorting nodes.The missing set, outside-exception set, and pairwise intersections are all empty.Source: §3.3.1 · pp. 6–7

3.3.2 Independent polynomial reconstruction

  • The checker reconstructs each cleared polynomial from its linear factors and checks every root with a complete degree-eleven coefficient array.Shared permutation prefixes reduce repeated work; the recurrence is also compared with direct expansion and exact point evaluations.Source: §3.3.2 · p. 7

3.3.3 Arithmetic safety and recorded workload

  • The untransformed polynomial coefficient norm is bounded by 2,921,591,507,056,640, below 2^52, supporting signed 64-bit classification arithmetic.Transformed-tree verification uses signed 128-bit arithmetic with safety checks, while the circuit inequality is checked with arbitrary-precision integers.Source: §3.3.3 · pp. 7–8
  • The ordinary batches check 81,961,459 circuits and 123,084,490,482 residual coefficient positions; sort-containing batches cover 36 roots with 4,575 leaves.The sort-containing batches contain 1, 29, and 6 roots, with 120, 3,533, and 922 leaves, respectively.Source: §3.3.3 · p. 8

3.4 Proof of the main theorem

The proof reduces distinct positive tuples to normalized rank words and nonnegative gaps, applies the finite certificates, then extends the result to ties by continuity.

3.4 Proof of the main theorem

  • For pairwise-distinct coordinates, cyclic rotation places the minimum first, yielding a unique rank word and strictly positive gaps.The cyclic expression is unchanged by rotation.Source: §3.4 · p. 8
  • The accepted class of each rank word proves its gap polynomial nonnegative, and the positive denominator then yields the inequality for every distinct positive tuple.The proof uses coefficientwise nonnegativity for one class, accepted identity-matrix leaves for another, and cover soundness for the tree classes.Source: §3.4 · p. 8
  • Distinct positive tuples are dense, so continuity extends the inequality to tuples with ties; when all coordinates equal a positive value, every numerator is zero.The denominator remains positive on the domain.Source: §3.4 · p. 8

4 Discussion

The discussion connects cumulative-gap coordinates, midpoint certificates, and orthant-covering substitutions to an exact finite proof over all normalized rank words.

4 Discussion

  • Cumulative sums encode ordered values as nonnegative parameters, while midpoint AM–GM circuits supply nonnegative polynomial summands.Exact expansion turns the certificate conditions into integer coefficient inequalities.Source: §4 · p. 8
  • The disjoint rank-word partition connects local decompositions on gap cones to every normalized rank word.The certificate conditions reduce to exact integer coefficient inequalities.Source: §4 · pp. 8–9
  • Binary comparison resolves one ordering relation at a time, while partial sorting handles a selected group at once; larger sorting steps are concentrated on 36 roots.Direct coefficients and ordinary midpoint leaves settle more than 99.9% of the root family.Source: §4 · p. 9
Loading 2610.04549v1…