Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Section III, p. 11, closing sentence (pinned Chebyshev form with (50)'s bound) — the fraction of instances below m/2 + (m/2) sin γ cos^{3D} γ − t is ⩽ m(6D+3)(D+1)/(4t²)

Proved
QAOAE3Lin2.Typical.typical_concentration

by mikedeng1 · Oct 15, 2026 · Mathlib 0df444a (Lean v4.33.1)

average-caseconcentrationmax-e3lin2p2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1qaoa

Fix any collection of mmm triples forming an E3LIN2 family (three distinct bits per triple, no triple repeated) in which every bit lies in at most D+1D+1D+1 triples. Fix γ\gammaγ with 0≤γ≤π/20\le\gamma\le\pi/20≤γ≤π/2 and t>0t>0t>0. Choose the right-hand sides of the mmm equations uniformly at random among the 2m2^m2m possibilities, run the level-one QAOA with angles (−γ,π/4)(-\gamma,\pi/4)(−γ,π/4), and let NNN be the expected number of satisfied equations of the measured string. Then

Pr⁡d[N≤m2+m2sin⁡γcos⁡3Dγ−t]  ≤  m(6D+3)(D+1)4t2.\Pr_d\Big[N\le\frac m2+\frac m2\sin\gamma\cos^{3D}\gamma-t\Big]\;\le\;\frac{m(6D+3)(D+1)}{4t^2}.dPr​[N≤2m​+2m​sinγcos3Dγ−t]≤4t2m(6D+3)(D+1)​.

With DDD fixed and t=cmt=c\sqrt mt=cm​, the right side is the constant (6D+3)(D+1)/(4c2)(6D+3)(D+1)/(4c^2)(6D+3)(D+1)/(4c2): for large mmm, on a typical instance, the QAOA satisfies in expectation at least (12+12sin⁡γcos⁡3Dγ)m\big(\tfrac12+\tfrac12\sin\gamma\cos^{3D}\gamma\big)m(21​+21​sinγcos3Dγ)m minus a term of order m\sqrt mm​ equations. At γ=1/3D\gamma=1/\sqrt{3D}γ=1/3D​ this threshold is (12+1+o(1)23eD1/2)m\big(\tfrac12+\tfrac{1+o(1)}{2\sqrt{3e}D^{1/2}}\big)m(21​+23e​D1/21+o(1)​)m.

Formalization Note The paper's sentence ("with high probability … at least (54) minus a term of order m\sqrt mm​") has no explicit failure probability. It is pinned here by combining the paper's own mean bound (50) and its variance bound 14m(6D+3)(D+1)\tfrac14m(6D+3)(D+1)41​m(6D+3)(D+1) through Chebyshev's inequality, which the page invokes implicitly. The threshold uses (50)'s exact lower bound m2sin⁡γcos⁡3Dγ\tfrac m2\sin\gamma\cos^{3D}\gamma2m​sinγcos3Dγ rather than (54), because (54) is false as a finite-DDD inequality (sin⁡γcos⁡3Dγ<1/3eD\sin\gamma\cos^{3D}\gamma<1/\sqrt{3eD}sinγcos3Dγ<1/3eD​ for every D≥1D\ge1D≥1). The range 0≤γ≤π/20\le\gamma\le\pi/20≤γ≤π/2 is added because (50)'s lower bound needs sin⁡γ,cos⁡γ≥0\sin\gamma,\cos\gamma\ge0sinγ,cosγ≥0; the paper's choice γ=1/3D\gamma=1/\sqrt{3D}γ=1/3D​ lies in it. The probability is over the right-hand sides only; NNN is the quantum expectation, not a single measurement outcome.

Preamble
import Mathlib
import Definitions.Def_QAOAE3Lin2_Typical_Setting
Formal statement
namespace QAOAE3Lin2.Typical

/-- §III, p. 11, closing sentence, pinned by Chebyshev's inequality from (50) and the variance
bound: for fixed triples with every bit in at most `D + 1` equations and `γ ∈ [0, π/2]`, the
fraction of the `2^m` right-hand-side choices on which the expected satisfied count of
`|−γ, π/4⟩` is at most `m/2 + (m/2) sin γ cos^{3D} γ − t` is at most `m(6D + 3)(D + 1)/(4t²)`. -/
theorem typical_concentration {n m : ℕ} (tri : Fin m → Fin n × Fin n × Fin n) (D : ℕ)
    (htri : QAOAE3Lin2.General.IsTripleFamily tri) (hocc : QAOAE3Lin2.General.BoundedOccurrence tri D) (γ : ℝ) (hγ₀ : 0 ≤ γ)
    (hγ₁ : γ ≤ Real.pi / 2) (t : ℝ) (ht : 0 < t) :
    ((Finset.univ.filter fun rhs : Fin m → Bool =>
        QAOAE3Lin2.General.expSat tri rhs (-γ) (Real.pi / 4) ≤
          (m : ℝ) / 2 + (m : ℝ) / 2 * Real.sin γ * Real.cos γ ^ (3 * D) - t).card : ℝ) / 2 ^ m ≤
      (m : ℝ) * (6 * D + 3) * (D + 1) / (4 * t ^ 2) := by sorry

end QAOAE3Lin2.Typical
Source
Farhi, Goldstone, Gutmann, A Quantum Approximate Optimization Algorithm Applied to a Bounded Occurrence Constraint Problem, arXiv:1412.6062v2, p. 11, Section III, closing paragraph (with (50), p. 10, and the variance bound, p. 11)
Human review
  • Endorsed by Shuze Chen · Oct 15, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 15, 2026

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

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 paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me