Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Exact limiting threshold of the single-cat family

Open
QAlgorithms.Papers.Independent.Support50.P40.cat_asymptoticThreshold

by Goku · Oct 23, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

For each positive block size p=m+1p=m+1p=m+1, let qp(f)q_p(f)qp​(f) be the signed yield of the ordinary cat code computed from its exact joint probabilities, and define τp=inf⁡{f∈[3/4,1]:qp(f)>0}\tau_p=\inf\{f\in[3/4,1]:q_p(f)>0\}τp​=inf{f∈[3/4,1]:qp​(f)>0}. These sets contain 111.

There is a unique f∞∈(3/4,1)f_\infty\in(3/4,1)f∞​∈(3/4,1) such that, for g∞=(1−f∞)/3g_\infty=(1-f_\infty)/3g∞​=(1−f∞​)/3,

(f∞−g∞)2f∞+g∞=8g∞(f∞+g∞),lim⁡p→∞τp=f∞.\frac{(f_\infty-g_\infty)^2}{f_\infty+g_\infty} =\sqrt{8g_\infty(f_\infty+g_\infty)},\qquad \lim_{p\to\infty}\tau_p=f_\infty.f∞​+g∞​(f∞​−g∞​)2​=8g∞​(f∞​+g∞​)​,p→∞lim​τp​=f∞​.

The limit includes both even and odd block sizes. This equation specifies the threshold exactly; .81808.81808.81808 is only the paper's numerical approximation. No convergence rate or exact version of the informal partial contributions in Eqs. (40)–(41) is asserted.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P40
Formal statement
namespace QAlgorithms.Papers.Independent.Support50.P40

open QAlgorithms QAlgorithms.Stabilizer

open scoped ComplexConjugate Matrix

noncomputable section

/-- PDF p.18, Eq.(42), Table I p.25, Fig.9 p.31: the high-fidelity
root is the limit of thresholds of the cat family, including both parities.
No exact decimal root, convergence rate, or intermediate asymptotic
identity from Eqs.(40)--(41) is asserted. -/
theorem cat_asymptoticThreshold :
    ∃ fstar : ℝ, (3 / 4 : ℝ) < fstar ∧ fstar < 1 ∧
      (fstar - noiseG fstar) ^ 2 / (fstar + noiseG fstar) =
        Real.sqrt (8 * noiseG fstar * (fstar + noiseG fstar)) ∧
      (∀ t : ℝ, (3 / 4 : ℝ) < t → t < 1 →
        (t - noiseG t) ^ 2 / (t + noiseG t) = Real.sqrt (8 * noiseG t * (t + noiseG t)) →
        t = fstar) ∧
      Filter.Tendsto catThreshold Filter.atTop (nhds fstar) := by sorry

end

end QAlgorithms.Papers.Independent.Support50.P40
Source
Quantum Channel Capacity of Very Noisy Channels, arXiv:quant-ph/9706061v3, PDF p. 18, Section V.A, Eq. (42): asymptotic cat-family threshold
Read-back

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

For each integer m≥0m\ge0m≥0 and real fff, put g(f)=(1−f)/3g(f)=(1-f)/3g(f)=(1−f)/3, and assign each error string e∈{I,X,Y,Z}m+1e\in\{I,X,Y,Z\}^{m+1}e∈{I,X,Y,Z}m+1 the weight fm+1−w(e)g(f)w(e)f^{m+1-w(e)}g(f)^{w(e)}fm+1−w(e)g(f)w(e), where w(e)w(e)w(e) counts its nonidentity entries. Associate to eee the syndrome s(e)q=a(eq+1)⊕a(e0)s(e)_q=a(e_{q+1})\mathbin{\oplus}a(e_0)s(e)q​=a(eq+1​)⊕a(e0​) and logical label L(e)L(e)L(e) determined by the pair (a(e0),⨁qb(eq))(a(e_0),\bigoplus_q b(e_q))(a(e0​),⨁q​b(eq​)), where the amplitude and phase pairs for I,X,Y,ZI,X,Y,ZI,X,Y,Z are respectively (0,0),(1,0),(1,1),(0,1)(0,0),(1,0),(1,1),(0,1)(0,0),(1,0),(1,1),(0,1). Let Jm,f(i,j)J_{m,f}(i,j)Jm,f​(i,j) be the sum of these weights over errors with syndrome iii and logical label jjj, let Sm,f(i)=∑jJm,f(i,j)S_{m,f}(i)=\sum_jJ_{m,f}(i,j)Sm,f​(i)=∑j​Jm,f​(i,j), and define

Ym(f)=1+H(Sm,f)−H(Jm,f)m+1,H(u)=−∑xuxlog⁡2ux,τm=inf⁡{f∈[3/4,1]:Ym(f)>0}.Y_m(f)=\frac{1+H(S_{m,f})-H(J_{m,f})}{m+1}, \qquad H(u)=-\sum_xu_x\log_2u_x, \qquad \tau_m=\inf\{f\in[3/4,1]:Y_m(f)>0\}.Ym​(f)=m+11+H(Sm,f​)−H(Jm,f​)​,H(u)=−x∑​ux​log2​ux​,τm​=inf{f∈[3/4,1]:Ym​(f)>0}.

Here log⁡20=0\log_2 0=0log2​0=0, and the real infimum of the empty set is assigned value 000. There exists a real f⋆f_\starf⋆​ with 3/4<f⋆<13/4<f_\star<13/4<f⋆​<1 such that

(f⋆−g(f⋆))2f⋆+g(f⋆)=8g(f⋆)(f⋆+g(f⋆)).\frac{(f_\star-g(f_\star))^2}{f_\star+g(f_\star)} =\sqrt{8g(f_\star)(f_\star+g(f_\star))}.f⋆​+g(f⋆​)(f⋆​−g(f⋆​))2​=8g(f⋆​)(f⋆​+g(f⋆​))​.

For every real ttt satisfying 3/4<t<13/4<t<13/4<t<1, if ttt satisfies this same equation with ttt in place of f⋆f_\starf⋆​, then t=f⋆t=f_\start=f⋆​; uniqueness is asserted within this open interval. Furthermore, τm→f⋆\tau_m\to f_\starτm​→f⋆​ as the natural number mmm tends to infinity: for every real ϵ>0\epsilon>0ϵ>0, there is an integer M≥0M\ge0M≥0 such that ∣τm−f⋆∣<ϵ|\tau_m-f_\star|<\epsilon∣τm​−f⋆​∣<ϵ for every integer m≥Mm\ge Mm≥M.

Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 24, 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