Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§2.3: degradable flagged identity/noise family (16)–(22)

Proved
QAlgorithms.Papers.Independent.Gap20.P19.flaggedIdentityDegradable

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

prove2quantum

All Hilbert spaces are finite dimensional, complex, and nonzero. A channel is completely positive and trace preserving on every operator. Write r(Φ)r(\Phi)r(Φ) for its Choi rank, the minimum pure-environment dimension. Receiver and complementary channels are always obtained from the same isometry.

Let N:Md→MdN:M_d\to M_dN:Md​→Md​ be given by an isometric dilation UUU, and let 1/2≤x≤11/2\le x\le11/2≤x≤1. Define

Fx,N(ρ)=x∣0⟩⟨0∣⊗ρ+(1−x)∣1⟩⟨1∣⊗N(ρ).F_{x,N}(\rho)=x|0\rangle\langle0|\otimes\rho+(1-x)|1\rangle\langle1|\otimes N(\rho).Fx,N​(ρ)=x∣0⟩⟨0∣⊗ρ+(1−x)∣1⟩⟨1∣⊗N(ρ).

Its normalized flag dilation has complement xTr⁡ρ⊕(1−x)NC(ρ)x\operatorname{Tr}\rho\oplus(1-x)N^C(\rho)xTrρ⊕(1−x)NC(ρ). On receiver blocks γ0,γ1\gamma_0,\gamma_1γ0​,γ1​, the CPTP degrader is

Dx(γ)=(2x−1xTr⁡γ0+Tr⁡γ1)⊕1−xxNC(γ0),D_x(\gamma)=\left(\frac{2x-1}{x}\operatorname{Tr}\gamma_0+\operatorname{Tr}\gamma_1\right)\oplus\frac{1-x}{x}N^C(\gamma_0),Dx​(γ)=(x2x−1​Trγ0​+Trγ1​)⊕x1−x​NC(γ0​),

with flag coherences discarded. Thus DxFx,N=Fx,NCD_xF_{x,N}=F_{x,N}^CDx​Fx,N​=Fx,NC​ and Fx,NF_{x,N}Fx,N​ is degradable. At x=1/2x=1/2x=1/2, its Choi rank is 1+r(N)1+r(N)1+r(N); for d>1d>1d>1 and r(N)=d2r(N)=d^2r(N)=d2, this exceeds output dimension 2d2d2d.

Scalar-first complement ordering is used consistently, including x=1x=1x=1.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Gap20_P19
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Gap20.P19

open scoped BigOperators Matrix ComplexConjugate ComplexOrder Kronecker

open Filter Topology

open QAlgorithms.Papers.Independent.Information100.P67
  (IsIsometry receiverChannel complementaryChannel IsDegradable IsAntidegradable
   maximumCoherentInformation dilationPower)

open QAlgorithms.Papers.Independent.Claimed50.P48
  (MinimalDilation indexedKrausDilation arbitraryVectorComplement superopTensor tensorDilation)

open QAlgorithms.Papers.Independent.Claimed50.P24 (ChannelEB)

open QAlgorithms.Papers.Independent.Information100.P60 (wernerHolevo)

open QAlgorithms.Papers.Independent.Gap48.P36 (finWeyl WeylLabel)

open QAlgorithms.Papers.Independent.Information100.P69 (choi)

open QAlgorithms.Papers.Independent.Information100.P50 (matrixRange)

open QAlgorithms.Papers.Independent.Information100.P08 (IsPure)

attribute [local instance] Classical.propDecidable Classical.decEq

variable {A B C E F R K A2 B2 E2 : Type}

variable [Fintype A] [DecidableEq A] [Nonempty A]

variable [Fintype B] [DecidableEq B] [Nonempty B]

variable [Fintype C] [DecidableEq C] [Nonempty C]

variable [Fintype E] [DecidableEq E] [Nonempty E]

variable [Fintype F] [DecidableEq F] [Nonempty F]

variable [Fintype R] [DecidableEq R] [Nonempty R]

variable [Fintype K] [DecidableEq K]

variable [Fintype A2] [DecidableEq A2] [Nonempty A2]

variable [Fintype B2] [DecidableEq B2] [Nonempty B2]

variable [Fintype E2] [DecidableEq E2] [Nonempty E2]

/-- Section 2.3, PDF pp.10–11, (16)–(22): corrected weights, normalized
common dilation, explicit degrader and minimal-rank enlargement at x=1/2.
At x=1 the noise Kraus operators vanish, so no 1+r_N rank assertion is made. -/
theorem flaggedIdentityDegradable [Nonempty A] [Nonempty E]
    (U : Matrix (A × E) A ℂ) (hU : IsIsometry U)
    (x : ℝ) (hx0 : 1/2 ≤ x) (hx1 : x ≤ 1) :
    receiverChannel (flaggedDilation x U) = flaggedIdentity x (receiverChannel U) ∧
    complementaryChannel (flaggedDilation x U) = flaggedComplement x U ∧
    IsDegradable (flaggedDilation x U) ∧ IsChannel (flaggedDegrader x U) ∧
    (∀ rho, flaggedDegrader x U (flaggedIdentity x (receiverChannel U) rho) =
      flaggedComplement x U rho) ∧
    (choi (flaggedIdentity (1/2) (receiverChannel U))).rank = 1+(choi (receiverChannel U)).rank ∧
    (1 < Fintype.card A → (choi (receiverChannel U)).rank = Fintype.card A^2 →
      2*Fintype.card A < (choi (flaggedIdentity (1/2) (receiverChannel U))).rank) := by sorry

end QAlgorithms.Papers.Independent.Gap20.P19

end
Source
The structure of degradable quantum channels, arXiv:0802.1360v2, PDF p. 10, §2.3: degradable flagged identity/noise family (16)–(22)
Read-back

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

Let A,EA,EA,E be nonempty finite types with decidable equality, let U:CA→CA⊗CEU:\mathbb C^A\to\mathbb C^A\otimes\mathbb C^EU:CA→CA⊗CE satisfy U†U=IU^\dagger U=IU†U=I, and let 1/2≤x≤11/2\le x\le11/2≤x≤1. Write N(X)=Tr⁡E(UXU†)N(X)=\operatorname{Tr}_E(UXU^\dagger)N(X)=TrE​(UXU†) and Nc(X)=Tr⁡A(UXU†)N^c(X)=\operatorname{Tr}_A(UXU^\dagger)Nc(X)=TrA​(UXU†). Define V:CA→C{false,true}×A⊗C{⊥}⊔EV:\mathbb C^A\to\mathbb C^{\{\mathrm{false},\mathrm{true}\}\times A}\otimes\mathbb C^{\{\bot\}\sqcup E}V:CA→C{false,true}×A⊗C{⊥}⊔E by

V((b,a),⊥),a′=x 1b=false, a=a′,V((b,a),e),a′=1−x 1b=trueU(a,e),a′.V_{((b,a),\bot),a'}=\sqrt{x}\,\mathbf1_{b=\mathrm{false},\,a=a'}, \qquad V_{((b,a),e),a'}=\sqrt{1-x}\,\mathbf1_{b=\mathrm{true}}U_{(a,e),a'}.V((b,a),⊥),a′​=x​1b=false,a=a′​,V((b,a),e),a′​=1−x​1b=true​U(a,e),a′​.

Its receiver map is Fx(X)=xX⊕(1−x)N(X)F_x(X)=xX\oplus(1-x)N(X)Fx​(X)=xX⊕(1−x)N(X), with the two blocks labelled false and true, and its environment map is Gx(X)=xTr⁡(X)⊕(1−x)Nc(X)G_x(X)=x\operatorname{Tr}(X)\oplus(1-x)N^c(X)Gx​(X)=xTr(X)⊕(1−x)Nc(X), with blocks labelled ⊥\bot⊥ and EEE. The matrix VVV is an isometry and its environment map is obtainable from its receiver map by a channel. More specifically, for any matrix YYY on the receiver, let Y0,Y1Y_0,Y_1Y0​,Y1​ be its false and true diagonal blocks; the map

Dx(Y)=(2x−1xTr⁡Y0+Tr⁡Y1)⊕1−xxNc(Y0)D_x(Y)= \left(\frac{2x-1}{x}\operatorname{Tr}Y_0+\operatorname{Tr}Y_1\right) \oplus\frac{1-x}{x}N^c(Y_0)Dx​(Y)=(x2x−1​TrY0​+TrY1​)⊕x1−x​Nc(Y0​)

has zero off-diagonal blocks, is a channel, and satisfies Dx(Fx(X))=Gx(X)D_x(F_x(X))=G_x(X)Dx​(Fx​(X))=Gx​(X) for every matrix XXX. A channel here means a finite Kraus map with ∑ℓKℓ†Kℓ=I\sum_\ell K_\ell^\dagger K_\ell=I∑ℓ​Kℓ†​Kℓ​=I. With J(Θ)(a,b),(a′,b′)=Θ(Eaa′)bb′J(\Theta)_{(a,b),(a',b')}=\Theta(E_{aa'})_{bb'}J(Θ)(a,b),(a′,b′)​=Θ(Eaa′​)bb′​, one also has rank⁡J(F1/2)=1+rank⁡J(N)\operatorname{rank}J(F_{1/2})=1+\operatorname{rank}J(N)rankJ(F1/2​)=1+rankJ(N); if ∣A∣>1|A|>1∣A∣>1 and rank⁡J(N)=∣A∣2\operatorname{rank}J(N)=|A|^2rankJ(N)=∣A∣2, then 2∣A∣<rank⁡J(F1/2)2|A|<\operatorname{rank}J(F_{1/2})2∣A∣<rankJ(F1/2​). Both endpoints of the allowed interval for xxx are included.

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