Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Exact cat diagonal QEC blocks and large-amplitude ratios

Open
QAlgorithms.Papers.Independent.Intakenext100.P65.cat_diagonalQec

by Gohan · Oct 31, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Fix integers S,ℓ≥0S,\ell\ge0S,ℓ≥0 and 0<γ<10<\gamma<10<γ<1. For real a>0a>0a>0, use the normalized projected-coherent cat basis, and set Nar=⟨a∣Πr∣a⟩N_a^r=\langle a|\Pi_r|a\rangleNar​=⟨a∣Πr​∣a⟩ with residues modulo 2(S+1)2(S+1)2(S+1). Put Pℓ=e−γa2(γa2)ℓ/ℓ!P_\ell=e^{-\gamma a^2}(\gamma a^2)^\ell/\ell!Pℓ​=e−γa2(γa2)ℓ/ℓ! and Rμ=Na1−γ(S+1)μ−ℓ/Na(S+1)μR_\mu=N_{a\sqrt{1-\gamma}}^{(S+1)\mu-\ell}/N_a^{(S+1)\mu}Rμ​=Na1−γ​(S+1)μ−ℓ​/Na(S+1)μ​. The actual compressed block has entries

⟨μa∣Eℓ†Eℓ∣ua⟩=δμuPℓRμ,cℓℓ=Pℓ2(R0+R1),zℓℓ=Pℓ2(R0−R1).\langle\mu_a|E_\ell^\dagger E_\ell| u_a\rangle=\delta_{\mu u}P_\ell R_\mu,\qquad c_{\ell\ell}=\frac{P_\ell}{2}(R_0+R_1),\qquad z_{\ell\ell}=\frac{P_\ell}{2}(R_0-R_1).⟨μa​∣Eℓ†​Eℓ​∣ua​⟩=δμu​Pℓ​Rμ​,cℓℓ​=2Pℓ​​(R0​+R1​),zℓℓ​=2Pℓ​​(R0​−R1​).

For each logical label, Rμ→1R_\mu\to1Rμ​→1 as a→∞a\to\inftya→∞. The comparison Poisson mean γa2\gamma a^2γa2 varies with amplitude, so this is not convergence to one fixed probability distribution.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P65
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P65

open MeasureTheory Filter

open QAlgorithms.Papers.Independent.Frontier18.P14 (displacementEntry)

open QAlgorithms.Papers.Independent.Dependency50.P40 (outputFirstChoi)

open scoped ComplexConjugate Matrix Topology

attribute [local instance] Classical.propDecidable

attribute [local instance] Classical.decEq

/-- PDF pp.10--11 (5.11)--(5.13): exact diagonal blocks and their two coefficients;
the large-a assertion is about ratios, not a fixed limiting Poisson law. -/
theorem cat_diagonalQec (S : ℕ) (g : ℝ) (hg0 : 0 < g) (hg1 : g < 1) (l : ℕ) :
    (∀ a > (0 : ℝ), qecBlock (catCodeCoordinates a S) g l l =
      fun i j => if i = j then ((poissonWeight a g l * catLossRatio a S g l i : ℝ) : ℂ) else 0) ∧
    (∀ a > (0 : ℝ),
      qecC (catCodeCoordinates a S) g l l = poissonWeight a g l *
        (catLossRatio a S g l 0 + catLossRatio a S g l 1) / 2 ∧
      qecZ (catCodeCoordinates a S) g l l = poissonWeight a g l *
        (catLossRatio a S g l 0 - catLossRatio a S g l 1) / 2) ∧
    (∀ mu : Fin 2, Tendsto (fun a => catLossRatio a S g l mu) atTop (nhds 1)) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P65

end
Source
Performance and structure of single-mode bosonic codes, arXiv:1708.05010v3, PDF p. 10, §V B (5.11)–(5.13): exact cat diagonal QEC blocks
Read-back

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

For every natural number SSS, real ggg with 0<g<10<g<10<g<1, and natural number lll, define, for real bbb and integer rrr,

C(b,r)=∑n=0∞∣1 n≡r  (mod 2(S+1))e−b2/2bnn!∣2,C(b,r)=\sum_{n=0}^{\infty} \left|\mathbf1_{\,n\equiv r\;(\mathrm{mod}\,2(S+1))} e^{-b^2/2}\frac{b^n}{\sqrt{n!}}\right|^2,C(b,r)=n=0∑∞​​1n≡r(mod2(S+1))​e−b2/2n!​bn​​2,

and, for a>0a>0a>0 and i∈{0,1}i\in\{0,1\}i∈{0,1}, set

vi,n=1 n≡(S+1)i  (mod 2(S+1))e−a2/2an/n!C(a,(S+1)i),wi,n=(n+ll)(g)l(1−g)nvi,n+l,v_{i,n} =\frac{\mathbf1_{\,n\equiv(S+1)i\;(\mathrm{mod}\,2(S+1))} e^{-a^2/2}a^n/\sqrt{n!}}{\sqrt{C(a,(S+1)i)}}, \quad w_{i,n}=\sqrt{\binom{n+l}{l}}(\sqrt g)^l(\sqrt{1-g})^n v_{i,n+l},vi,n​=C(a,(S+1)i)​1n≡(S+1)i(mod2(S+1))​e−a2/2an/n!​​,wi,n​=(ln+l​)​(g​)l(1−g​)nvi,n+l​, Ri(a)=C(a1−g,(S+1)i−l)C(a,(S+1)i),pl(a)=(ga2)le−ga2l!,Qij(a)=∑n=0∞wi,n‾wj,n.R_i(a)=\frac{C(a\sqrt{1-g},(S+1)i-l)}{C(a,(S+1)i)}, \qquad p_l(a)=\frac{(ga^2)^l e^{-ga^2}}{l!}, \qquad Q_{ij}(a)=\sum_{n=0}^{\infty}\overline{w_{i,n}}w_{j,n}.Ri​(a)=C(a,(S+1)i)C(a1−g​,(S+1)i−l)​,pl​(a)=l!(ga2)le−ga2​,Qij​(a)=n=0∑∞​wi,n​​wj,n​.

The residue difference in RiR_iRi​ is taken in Z\mathbb ZZ, including negative values. The theorem asserts that, for every a>0a>0a>0,

Qij(a)={pl(a)Ri(a),i=j,0,i≠j,Q_{ij}(a)= \begin{cases} p_l(a)R_i(a),&i=j,\\ 0,&i\ne j, \end{cases}Qij​(a)={pl​(a)Ri​(a),0,​i=j,i=j,​

and also

Re⁡Q00(a)+Q11(a)2=pl(a)R0(a)+R1(a)2,Re⁡Q00(a)−Q11(a)2=pl(a)R0(a)−R1(a)2.\operatorname{Re}\frac{Q_{00}(a)+Q_{11}(a)}2 =p_l(a)\frac{R_0(a)+R_1(a)}2, \qquad \operatorname{Re}\frac{Q_{00}(a)-Q_{11}(a)}2 =p_l(a)\frac{R_0(a)-R_1(a)}2.Re2Q00​(a)+Q11​(a)​=pl​(a)2R0​(a)+R1​(a)​,Re2Q00​(a)−Q11​(a)​=pl​(a)2R0​(a)−R1​(a)​.

Finally, for each i∈{0,1}i\in\{0,1\}i∈{0,1}, it asserts lim⁡a→+∞Ri(a)=1\lim_{a\to+\infty}R_i(a)=1lima→+∞​Ri​(a)=1. There is no upper bound on lll, and S=0S=0S=0 and l=0l=0l=0 are included.

Human review
  • Endorsed by Shuze Chen · Oct 31, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 31, 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