Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 4: additivity of maximum coherent information for degradable channels

Open
QAlgorithms.Papers.Independent.Support50.P07.coherentInformation_additive

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

prove2quantum

Let N1\mathcal N_1N1​ and N2\mathcal N_2N2​ be any two degradable channels, with independently fixed Stinespring isometries. For each channel write Q(1)Q^{(1)}Q(1) for the supremum of S(B)−S(E)S(B)-S(E)S(B)−S(E) over all normalized input density operators. Then

Q(1)(N1⊗N2)=Q(1)(N1)+Q(1)(N2).Q^{(1)}(\mathcal N_1\otimes\mathcal N_2)=Q^{(1)}(\mathcal N_1)+Q^{(1)}(\mathcal N_2).Q(1)(N1​⊗N2​)=Q(1)(N1​)+Q(1)(N2​).

The product optimization allows entangled inputs.

All quantum registers are finite and nonzero, states have trace one, pure vectors have norm one, and entropy and logarithms are in bits. Operational private capacities use memoryless powers, positive blocklengths and message counts, uniform messages, arbitrary mixed block inputs, and normalized decoding POVMs. A code is good when its joint decoded message/environment state has full trace norm less than the error from a perfectly correlated uniform pair tensor its actual common environment marginal. Rates use every positive error and rate slack at all sufficiently large blocklengths. Symmetric assistance allows one arbitrary finite symmetric channel per block without charging its dimension.

This result supplies a capacity identity or bound for the channel family and resource model stated above.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P07
Formal statement
noncomputable section

open scoped BigOperators Matrix ComplexConjugate ComplexOrder

open Filter

namespace QAlgorithms.Papers.Independent.Support50.P07

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

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

open QAlgorithms.Papers.Independent.Dependency50.P19 (mutualInformation)

variable {A B E C D F : Type}

variable [Fintype A] [DecidableEq A] [Fintype B] [DecidableEq B]

variable [Fintype E] [DecidableEq E] [Fintype C] [DecidableEq C]

variable [Fintype D] [DecidableEq D] [Fintype F] [DecidableEq F]

/-- Lemma 4, PDF p.3, prior result [9] with an alternate proof here: additivity for
TWO arbitrary degradable channels, without restricting the joint input to product states. -/
theorem coherentInformation_additive [Nonempty A] [Nonempty B] [Nonempty E]
    [Nonempty C] [Nonempty D] [Nonempty F]
    (U : Matrix (B × E) A ℂ)
    (V : Matrix (D × F) C ℂ) (hU : IsIsometry U) (hV : IsIsometry V)
    (hdU : IsDegradable U) (hdV : IsDegradable V) :
    maximumCoherentInformation (productDilation U V) =
      maximumCoherentInformation U + maximumCoherentInformation V := by sorry

end QAlgorithms.Papers.Independent.Support50.P07

end
Source
The private classical capacity with a symmetric side channel and its application to quantum cryptography, arXiv:0705.3838v1, PDF p. 3, Lemma 4: additivity of maximum coherent information for degradable channels
Read-back

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

Let A,B,E,C,D,FA,B,E,C,D,FA,B,E,C,D,F be arbitrary finite nonempty index types with decidable equality. Let U:A→B×EU:A\to B\times EU:A→B×E and V:C→D×FV:C\to D\times FV:C→D×F be complex matrices satisfying U†U=IAU^\dagger U=I_AU†U=IA​ and V†V=ICV^\dagger V=I_CV†V=IC​, and assume each is degradable. Explicitly, writing

NU(ρ)=tr⁡E(UρU†),NUc(ρ)=tr⁡B(UρU†),N_U(\rho)=\operatorname{tr}_E(U\rho U^\dagger),\qquad N_U^c(\rho)=\operatorname{tr}_B(U\rho U^\dagger),NU​(ρ)=trE​(UρU†),NUc​(ρ)=trB​(UρU†),

and defining NV,NVcN_V,N_V^cNV​,NVc​ analogously, the degradability assumptions include the isometry conditions and maps TU,TVT_U,T_VTU​,TV​ satisfying TU(NU(ρ))=NUc(ρ)T_U(N_U(\rho))=N_U^c(\rho)TU​(NU​(ρ))=NUc​(ρ) and TV(NV(σ))=NVc(σ)T_V(N_V(\sigma))=N_V^c(\sigma)TV​(NV​(σ))=NVc​(σ) for every input matrix. Each such map must have a finite Kraus representation T(X)=∑jKjXKj†T(X)=\sum_jK_jXK_j^\daggerT(X)=∑j​Kj​XKj†​ with ∑jKj†Kj=I\sum_jK_j^\dagger K_j=I∑j​Kj†​Kj​=I. Define W:A×C→(B×D)×(E×F)W:A\times C\to(B\times D)\times(E\times F)W:A×C→(B×D)×(E×F) by

W((b,d),(e,f)),(a,c)=U(b,e),aV(d,f),c.W_{((b,d),(e,f)),(a,c)}=U_{(b,e),a}V_{(d,f),c}.W((b,d),(e,f)),(a,c)​=U(b,e),a​V(d,f),c​.

Then

sup⁡ρ≥0, tr⁡ρ=1[S ⁣(tr⁡E×F(WρW†))−S ⁣(tr⁡B×D(WρW†))]=sup⁡α≥0, tr⁡α=1[S(NU(α))−S(NUc(α))]+sup⁡β≥0, tr⁡β=1[S(NV(β))−S(NVc(β))],\begin{aligned} &\sup_{\rho\ge0,\ \operatorname{tr}\rho=1} \left[S\!\left(\operatorname{tr}_{E\times F}(W\rho W^\dagger)\right) -S\!\left(\operatorname{tr}_{B\times D}(W\rho W^\dagger)\right)\right]\\ &\quad= \sup_{\alpha\ge0,\ \operatorname{tr}\alpha=1} \bigl[S(N_U(\alpha))-S(N_U^c(\alpha))\bigr] + \sup_{\beta\ge0,\ \operatorname{tr}\beta=1} \bigl[S(N_V(\beta))-S(N_V^c(\beta))\bigr], \end{aligned}​ρ≥0, trρ=1sup​[S(trE×F​(WρW†))−S(trB×D​(WρW†))]=α≥0, trα=1sup​[S(NU​(α))−S(NUc​(α))]+β≥0, trβ=1sup​[S(NV​(β))−S(NVc​(β))],​

where ρ,α,β\rho,\alpha,\betaρ,α,β are matrices on A×C,A,CA\times C,A,CA×C,A,C, respectively, and S(X)=−Re⁡tr⁡f(X)S(X)=-\operatorname{Re}\operatorname{tr}f(X)S(X)=−Retrf(X), with f(x)=xlog⁡2xf(x)=x\log_2xf(x)=xlog2​x, f(0)=0f(0)=0f(0)=0, evaluated by continuous functional calculus. The joint supremum permits every density matrix on A×CA\times CA×C, including those that are not product matrices.

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