Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 16

Open
QAlgorithms.Papers.Independent.Gap50repair.P01.rotation_generalized_spectral_formula

by Goku · Oct 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let n≥1n\ge1n≥1 and A∈Sp⁡(2n,R)A\in\operatorname{Sp}(2n,\mathbb R)A∈Sp(2n,R). Let EλE_\lambdaEλ​ be its full complex generalized eigenspace. For every eigenvalue λ\lambdaλ on the unit circle other than ±1\pm1±1, the real bilinear form

Qλ(v,w)=Im⁡Ω0(v,w‾),v,w∈Eλ,Q_\lambda(v,w)=\operatorname{Im}\Omega_0(v,\overline w),\qquad v,w\in E_\lambda,Qλ​(v,w)=ImΩ0​(v,w),v,w∈Eλ​,

is symmetric and nondegenerate on the underlying real space. Its positive inertia m+(λ)m^+(\lambda)m+(λ) is even. Write m−=∑λ∈R, λ<0dim⁡CEλm^-=\sum_{\lambda\in\mathbb R,\,\lambda<0}\dim_{\mathbb C}E_\lambdam−=∑λ∈R,λ<0​dimC​Eλ​; this integer is also even. The spectral rotation map satisfies

ρ(A)=(−1)m−/2∏λ∈σ(A)∩S1λ≠±1λm+(λ)/2.\rho(A)=(-1)^{m^-/2}\prod_{\substack{\lambda\in\sigma(A)\cap S^1\\\lambda\ne\pm1}}\lambda^{m^+(\lambda)/2}.ρ(A)=(−1)m−/2λ∈σ(A)∩S1λ=±1​∏​λm+(λ)/2.

The product ranges over distinct eigenvalues, with their contributions weighted by these multiplicities. All powers have integer exponents. This formula determines the rotation contribution of each spectral component without choosing fractional-power branches.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Gap50repair_P01
Formal statement
noncomputable section

open scoped BigOperators Matrix Matrix.Norms.Frobenius

namespace QAlgorithms.Papers.Independent.Gap50repair.P01

attribute [local instance] Classical.propDecidable

-- Pinned entrywise complexification and symplecticity. The latter uses -Omega0;
-- preserving -Omega0 is equivalent to preserving Omega0, with identical coordinates.

open QAlgorithms.Papers.Independent.Dependency50.P10 (complexify)

open QAlgorithms.Papers.Independent.Frontier30.P02 (IsSymplectic)

/-- Theorem 16 ([12,3]), PDF pp.14–15: real symmetric nondegenerate Krein
forms on the full generalized eigenspaces and formula (18). The half exponents
are integers; no fractional complex power is used. -/
theorem rotation_generalized_spectral_formula {n : ℕ} (hn : 1 ≤ n)
    (A : GroupedRealMatrix n) (hA : IsSymplectic A) :
    (∀ z ∈ eigenvalues A, ‖z‖ = 1 → z ≠ 1 → z ≠ -1 →
      (∀ v ∈ generalizedSpace A z, ∀ w ∈ generalizedSpace A z,
        kreinPairing v w = kreinPairing w v) ∧
      (∀ v ∈ generalizedSpace A z,
        (∀ w ∈ generalizedSpace A z, kreinPairing v w = 0) → v = 0) ∧
      2 ∣ kreinPositive A z) ∧
    (2 ∣ ∑ z ∈ (eigenvalues A).filter (fun z => z.im = 0 ∧ z.re < 0),
      Module.finrank ℂ (generalizedSpace A z)) ∧
    spectralRotation A =
      (-1 : ℂ)^((∑ z ∈ (eigenvalues A).filter (fun z => z.im = 0 ∧ z.re < 0),
        Module.finrank ℂ (generalizedSpace A z))/2) *
      ∏ z ∈ (eigenvalues A).filter (fun z => ‖z‖ = 1 ∧ z ≠ 1 ∧ z ≠ -1),
        z^(kreinPositive A z/2) := by sorry

end QAlgorithms.Papers.Independent.Gap50repair.P01

end
Source
The Conley-Zehnder index for a path of symplectic matrices, arXiv:1201.3728v2, PDF p. 14, Theorem 16
Read-back

What the Lean code literally says, in plain math · gpt-6.1-sol

For every natural number n≥1n\geq1n≥1 and every real 2n×2n2n\times2n2n×2n matrix AAA satisfying ATJnA=JnA^{\mathsf T}J_nA=J_nATJn​A=Jn​, where Jn=(0−InIn0)J_n=\begin{pmatrix}0&-I_n\\I_n&0\end{pmatrix}Jn​=(0In​​−In​0​), let E(A)E(A)E(A) be the finite set of distinct roots of the complex characteristic polynomial of AAA, let aA(z)a_A(z)aA​(z) denote the root multiplicity, and put VA(z)=ker⁡(AC−zI)2nV_A(z)=\ker(A_{\mathbb C}-zI)^{2n}VA​(z)=ker(AC​−zI)2n. On complex vectors define the real pairing

K(v,w)=Im⁡∑i=1n(vx,iwy,i‾−vy,iwx,i‾),K(v,w)=\operatorname{Im}\sum_{i=1}^n \bigl(v_{x,i}\overline{w_{y,i}}-v_{y,i}\overline{w_{x,i}}\bigr),K(v,w)=Imi=1∑n​(vx,i​wy,i​​−vy,i​wx,i​​),

and let κA(z)\kappa_A(z)κA​(z) be the maximum k∈{0,…,4n}k\in\{0,\ldots,4n\}k∈{0,…,4n} for which some real subspace W⊆VA(z)W\subseteq V_A(z)W⊆VA​(z) has real dimension kkk and K(v,v)>0K(v,v)>0K(v,v)>0 for every nonzero v∈Wv\in Wv∈W. The theorem asserts that, for every z∈E(A)z\in E(A)z∈E(A) with ∣z∣=1|z|=1∣z∣=1, z≠1z\neq1z=1, and z≠−1z\neq-1z=−1, the pairing is symmetric on VA(z)V_A(z)VA​(z), meaning K(v,w)=K(w,v)K(v,w)=K(w,v)K(v,w)=K(w,v) for all v,w∈VA(z)v,w\in V_A(z)v,w∈VA​(z); it has zero left kernel there, meaning that v∈VA(z)v\in V_A(z)v∈VA​(z) and K(v,w)=0K(v,w)=0K(v,w)=0 for every w∈VA(z)w\in V_A(z)w∈VA​(z) imply v=0v=0v=0; and 222 divides κA(z)\kappa_A(z)κA​(z). It also asserts that 222 divides ∑z∈E(A), Im⁡z=0, Re⁡z<0dim⁡CVA(z)\sum_{z\in E(A),\ \operatorname{Im}z=0,\ \operatorname{Re}z<0}\dim_{\mathbb C}V_A(z)∑z∈E(A), Imz=0, Rez<0​dimC​VA​(z), and that the quantity

ρ(A):=(−1)⌊∑z∈E(A), Im⁡z=0, Re⁡z<0aA(z)2⌋∏z∈E(A)∣z∣=1, z≠1, z≠−1z⌊κA(z)/2⌋\rho(A):= (-1)^{\left\lfloor\frac{\sum_{z\in E(A),\ \operatorname{Im}z=0,\ \operatorname{Re}z<0}a_A(z)}2\right\rfloor} \prod_{\substack{z\in E(A)\\|z|=1,\ z\neq1,\ z\neq-1}} z^{\lfloor\kappa_A(z)/2\rfloor}ρ(A):=(−1)⌊2∑z∈E(A), Imz=0, Rez<0​aA​(z)​⌋z∈E(A)∣z∣=1, z=1, z=−1​∏​z⌊κA​(z)/2⌋

satisfies

ρ(A)=(−1)∑z∈E(A), Im⁡z=0, Re⁡z<0dim⁡CVA(z)2∏z∈E(A)∣z∣=1, z≠1, z≠−1zκA(z)/2.\rho(A)= (-1)^{\frac{\sum_{z\in E(A),\ \operatorname{Im}z=0,\ \operatorname{Re}z<0}\dim_{\mathbb C}V_A(z)}2} \prod_{\substack{z\in E(A)\\|z|=1,\ z\neq1,\ z\neq-1}} z^{\kappa_A(z)/2}.ρ(A)=(−1)2∑z∈E(A), Imz=0, Rez<0​dimC​VA​(z)​z∈E(A)∣z∣=1, z=1, z=−1​∏​zκA​(z)/2.

Every sum and product is over distinct roots; empty sums and products are 000 and 111. Division in all exponents is natural-number division, with the asserted divisibility making the quotients on the right exact.

Human review
  • Endorsed by Shuze Chen · Oct 22, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 22, 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