Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Coset masses identify syndromes and residual logical Paulis

Proved
QAlgorithms.Papers.Independent.Support50.P40.coset_probabilities

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

prove2quantum

Fix a normalized additive code encoding one qubit into p≥1p\ge1p≥1 physical qubits. Its phase-inclusive stabilizer is commuting, excludes −I-I−I, and has p−1p-1p−1 independent generators. Let Sˉ\bar SSˉ be its phase-free label set and Sˉ⊥\bar S^\perpSˉ⊥ its commutation centralizer. Choose complete representatives eie_iei​ of physical syndromes and sjs_jsj​ of logical j∈{I,X,Y,Z}j\in\{I,X,Y,Z\}j∈{I,X,Y,Z}, with the identity representatives at the zero syndrome and logical identity. The logical representatives act as the named Paulis on the entire encoded operator algebra. Let PfP_fPf​ be the independent depolarizing error law, with identity probability f∈[0,1]f\in[0,1]f∈[0,1] and other-Pauli probability (1−f)/3(1-f)/3(1−f)/3.

Then Pf(Sˉ)P_f(\bar S)Pf​(Sˉ) is exactly the mass of errors acting as the identity channel on the whole code. For j≠Ij\ne Ij=I, Pf(sjSˉ)P_f(s_j\bar S)Pf​(sj​Sˉ) is the mass of undetected errors acting as logical jjj; Pf(Sˉ⊥)P_f(\bar S^\perp)Pf​(Sˉ⊥) is the no-syndrome probability. For every syndrome and logical label,

ai=Pr⁡(i)=Pf(eiSˉ⊥),bij=Pr⁡(i,j)=Pf(eisjSˉ).a_i=\Pr(i)=P_f(e_i\bar S^\perp),\qquad b_{ij}=\Pr(i,j)=P_f(e_i s_j\bar S).ai​=Pr(i)=Pf​(ei​Sˉ⊥),bij​=Pr(i,j)=Pf​(ei​sj​Sˉ).

For the normalized encoded Bell input these are the syndrome and generalized Bell measurement probabilities. After correcting with ei−1e_i^{-1}ei−1​, the unnormalized logical branch on any logical operator ρ\rhoρ is ∑jbij σjρσj†\sum_j b_{ij}\,\sigma_j\rho\sigma_j^\dagger∑j​bij​σj​ρσj†​. Logical I,X,Y,ZI,X,Y,ZI,X,Y,Z correspond to Φ+,Ψ+,Ψ−,Φ−\Phi^+,\Psi^+,\Psi^-,\Phi^-Φ+,Ψ+,Ψ−,Φ− as density operators.

These identifications retain zero-mass outcomes without conditional division and supply the probability law used in the concatenation rate.

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 pp.7--8,13, five probability interpretations and Eqs.(18)--(19).
All logical actions are compared on the entire encoded operator algebra;
the recovered branch retains its mass even for zero-probability syndromes. -/
theorem coset_probabilities {m : ℕ} (C : InnerCode m) (T : CosetChoices C)
    (f : ℝ) (hf0 : 0 ≤ f) (hf1 : f ≤ 1) :
    weightPolynomial f (stabilizerLabels C.stabilizer) =
      errorEventMass f (fun e : ErrorLabel (m + 1) =>
        ∀ rho : Matrix Bool Bool ℂ,
          conjSuperop (pauliOps e) (conjSuperop (innerEncoderMatrix C) rho) =
            conjSuperop (innerEncoderMatrix C) rho) ∧
    (∀ j : Pauli, j ≠ .I →
      weightPolynomial f (labelCoset (T.logicalRep j) (stabilizerLabels C.stabilizer)) =
        errorEventMass f (fun e : ErrorLabel (m + 1) =>
          e ∈ centralizerLabels C.stabilizer ∧ ∀ rho : Matrix Bool Bool ℂ,
            conjSuperop (pauliOps e) (conjSuperop (innerEncoderMatrix C) rho) =
              conjSuperop (innerEncoderMatrix C) (conjSuperop j.mat rho))) ∧
    weightPolynomial f (centralizerLabels C.stabilizer) =
      effectProb (syndromeProjector C T (fun _ => false)) (outputQ C f) ∧
    (∀ i, syndromeMass C T f i =
      effectProb (syndromeProjector C T i) (outputQ C f)) ∧
    (∀ i j, jointMass C T f i j =
      effectProb (pureDensity (bellBasis C T i j)) (outputRQ C f)) ∧
    (∀ i (rho : Matrix Bool Bool ℂ), recoveredBranch C T f i rho =
      ∑ j : Pauli, (jointMass C T f i j : ℂ) • conjSuperop j.mat rho) := by sorry

end

end QAlgorithms.Papers.Independent.Support50.P40
Source
Quantum Channel Capacity of Very Noisy Channels, arXiv:quant-ph/9706061v3, PDF p. 8, Section II bullets and Eqs. (18)-(19): coset probabilities and logical outcomes
Read-back

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

Let mmm be any natural number, put n=m+1n=m+1n=m+1, and write Qr={0,1}rQ_r=\{0,1\}^rQr​={0,1}r. Use I,X,Y,ZI,X,Y,ZI,X,Y,Z for the usual Pauli matrices in the Boolean basis, with Y0,1=−iY_{0,1}=-\mathrm iY0,1​=−i, and let Pe=⨂q=0n−1eqP_e=\bigotimes_{q=0}^{n-1}e_qPe​=⨂q=0n−1​eq​ for each Pauli label string eee. Multiplication of labels, denoted by ⊙\odot⊙, discards the phase in Pauli multiplication: under I↦(0,0)I\mapsto(0,0)I↦(0,0), X↦(1,0)X\mapsto(1,0)X↦(1,0), Y↦(1,1)Y\mapsto(1,1)Y↦(1,1), Z↦(0,1)Z\mapsto(0,1)Z↦(0,1), it is coordinatewise addition modulo 222. The supplied inner-code data consist of a subgroup SSS of the unitary matrices on CQn\mathbb C^{Q_n}CQn​, contained in the subgroup generated by all matrices icPe\mathrm i^cP_eicPe​ with c∈{0,1,2,3}c\in\{0,1,2,3\}c∈{0,1,2,3}, whose elements commute pairwise and none of which is −I-I−I; and two orthonormal vectors v0,v1v_0,v_1v0​,v1​ whose span is exactly the space fixed by every element of SSS. Let E:C{0,1}→CQnE:\mathbb C^{\{0,1\}}\to\mathbb C^{Q_n}E:C{0,1}→CQn​ have columns v0,v1v_0,v_1v0​,v1​, and define

A={e:∃U∈S, ∃c∈{0,1,2,3}, Pe=icU},B={e:∀U∈S, PeU=UPe}.A=\{e:\exists U\in S,\ \exists c\in\{0,1,2,3\},\ P_e=\mathrm i^cU\}, \qquad B=\{e:\forall U\in S,\ P_eU=UP_e\}.A={e:∃U∈S, ∃c∈{0,1,2,3}, Pe​=icU},B={e:∀U∈S, Pe​U=UPe​}.

The inner-code data also require ∣A∣=2m|A|=2^m∣A∣=2m. The supplied representative data consist of strings sis_isi​ for i∈Qmi\in Q_mi∈Qm​ and ℓj\ell_jℓj​ for j∈{I,X,Y,Z}j\in\{I,X,Y,Z\}j∈{I,X,Y,Z}, with s0=ℓI=(I,…,I)s_0=\ell_I=(I,\ldots,I)s0​=ℓI​=(I,…,I); every label string belongs to si⊙B={si⊙e:e∈B}s_i\odot B=\{s_i\odot e:e\in B\}si​⊙B={si​⊙e:e∈B} for exactly one iii; every ℓj\ell_jℓj​ belongs to BBB; every element of BBB belongs to ℓj⊙A\ell_j\odot Aℓj​⊙A for exactly one jjj; and, for every jjj and every complex 2×22\times22×2 matrix ρ\rhoρ,

PℓjEρE†Pℓj†=E jρj†E†.P_{\ell_j}E\rho E^\dagger P_{\ell_j}^\dagger =E\,j\rho j^\dagger E^\dagger.Pℓj​​EρE†Pℓj​†​=Ejρj†E†.

For every real fff with 0≤f≤10\le f\le10≤f≤1, let w(e)w(e)w(e) count nonidentity entries, set νf(e)=fn−w(e)((1−f)/3)w(e)\nu_f(e)=f^{n-w(e)}((1-f)/3)^{w(e)}νf​(e)=fn−w(e)((1−f)/3)w(e), and write Wf(D)=∑e∈Dνf(e)W_f(D)=\sum_{e\in D}\nu_f(e)Wf​(D)=∑e∈D​νf​(e). Define

sf(i)=Wf(si⊙B),jf(i,j)=Wf((si⊙ℓj)⊙A),Γf(X)=∑eνf(e)PeXPe†,∣Ω⟩=∣0⟩⊗v0+∣1⟩⊗v12,ρRQ=(id⊗Γf)(∣Ω⟩⟨Ω∣),ρQ=tr⁡RρRQ,ui,b=Psivb,Πi=∑b∈{0,1}∣ui,b⟩⟨ui,b∣,βi,j=(I⊗Psi⊙ℓj)∣Ω⟩.\begin{gathered} s_f(i)=W_f(s_i\odot B),\qquad j_f(i,j)=W_f((s_i\odot\ell_j)\odot A),\\ \Gamma_f(X)=\sum_e\nu_f(e)P_eXP_e^\dagger,\qquad |\Omega\rangle=\frac{|0\rangle\otimes v_0+|1\rangle\otimes v_1}{\sqrt2},\\ \rho_{RQ}=(\mathrm{id}\otimes\Gamma_f)(|\Omega\rangle\langle\Omega|), \qquad \rho_Q=\operatorname{tr}_{R}\rho_{RQ},\\ u_{i,b}=P_{s_i}v_b,\qquad \Pi_i=\sum_{b\in\{0,1\}}|u_{i,b}\rangle\langle u_{i,b}|,\qquad \beta_{i,j}=(I\otimes P_{s_i\odot\ell_j})|\Omega\rangle . \end{gathered}sf​(i)=Wf​(si​⊙B),jf​(i,j)=Wf​((si​⊙ℓj​)⊙A),Γf​(X)=e∑​νf​(e)Pe​XPe†​,∣Ω⟩=2​∣0⟩⊗v0​+∣1⟩⊗v1​​,ρRQ​=(id⊗Γf​)(∣Ω⟩⟨Ω∣),ρQ​=trR​ρRQ​,ui,b​=Psi​​vb​,Πi​=b∈{0,1}∑​∣ui,b​⟩⟨ui,b​∣,βi,j​=(I⊗Psi​⊙ℓj​​)∣Ω⟩.​

Here id⊗Γf\mathrm{id}\otimes\Gamma_fid⊗Γf​ applies Γf\Gamma_fΓf​ to each block indexed by the reference bit, and (ρQ)x,y=∑b∈{0,1}(ρRQ)(b,x),(b,y)(\rho_Q)_{x,y}=\sum_{b\in\{0,1\}}(\rho_{RQ})_{(b,x),(b,y)}(ρQ​)x,y​=∑b∈{0,1}​(ρRQ​)(b,x),(b,y)​. The theorem asserts the conjunction

Wf(A)=∑e: ∀ρ∈M2(C),PeEρE†Pe†=EρE†νf(e),Wf(ℓj⊙A)=∑e∈B: ∀ρ∈M2(C),PeEρE†Pe†=Ejρj†E†νf(e)for every j≠I,Wf(B)=Re⁡tr⁡(Π0ρQ),sf(i)=Re⁡tr⁡(ΠiρQ)for every i∈Qm,jf(i,j)=Re⁡tr⁡(∣βi,j⟩⟨βi,j∣ρRQ)for every i∈Qm, j∈{I,X,Y,Z},E†Psi†Γf(EρE†)PsiE=∑j∈{I,X,Y,Z}jf(i,j) jρj†for every i∈Qm, ρ∈M2(C).\begin{aligned} W_f(A) &=\sum_{\substack{e:\ \forall\rho\in M_2(\mathbb C),\\ P_eE\rho E^\dagger P_e^\dagger=E\rho E^\dagger}}\nu_f(e),\\ W_f(\ell_j\odot A) &=\sum_{\substack{e\in B:\ \forall\rho\in M_2(\mathbb C),\\ P_eE\rho E^\dagger P_e^\dagger=E\,j\rho j^\dagger E^\dagger}}\nu_f(e) &&\text{for every }j\ne I,\\ W_f(B)&=\operatorname{Re}\operatorname{tr}(\Pi_0\rho_Q),\\ s_f(i)&=\operatorname{Re}\operatorname{tr}(\Pi_i\rho_Q) &&\text{for every }i\in Q_m,\\ j_f(i,j)&=\operatorname{Re}\operatorname{tr} \bigl(|\beta_{i,j}\rangle\langle\beta_{i,j}|\rho_{RQ}\bigr) &&\text{for every }i\in Q_m,\ j\in\{I,X,Y,Z\},\\ E^\dagger P_{s_i}^\dagger\Gamma_f(E\rho E^\dagger)P_{s_i}E &=\sum_{j\in\{I,X,Y,Z\}}j_f(i,j)\,j\rho j^\dagger &&\text{for every }i\in Q_m,\ \rho\in M_2(\mathbb C). \end{aligned}Wf​(A)Wf​(ℓj​⊙A)Wf​(B)sf​(i)jf​(i,j)E†Psi​†​Γf​(EρE†)Psi​​E​=e: ∀ρ∈M2​(C),Pe​EρE†Pe†​=EρE†​∑​νf​(e),=e∈B: ∀ρ∈M2​(C),Pe​EρE†Pe†​=Ejρj†E†​∑​νf​(e)=Retr(Π0​ρQ​),=Retr(Πi​ρQ​)=Retr(∣βi,j​⟩⟨βi,j​∣ρRQ​)=j∈{I,X,Y,Z}∑​jf​(i,j)jρj†​​for every j=I,for every i∈Qm​,for every i∈Qm​, j∈{I,X,Y,Z},for every i∈Qm​, ρ∈M2​(C).​

The matrices ρ\rhoρ quantified in the action conditions and final equality are arbitrary complex matrices, without positivity or trace assumptions. The theorem includes m=0m=0m=0 and f=0,1f=0,1f=0,1, and all powers with exponent 000 equal 111.

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