Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 1: four equivalent additivity properties

Open
QAlgorithms.Papers.Independent.Gap48.P46.additivity_equivalences

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

prove2quantum

All quantum systems are nonzero finite-dimensional complex Hilbert spaces. States are positive semidefinite trace-one matrices, and channels are CPTP maps on the full matrix algebra. All entropies and logarithms are in bits. An ensemble is a finite probability distribution of states, with zero weights permitted. For a channel Φ\PhiΦ, write χΦ(ρ)\chi_\Phi(\rho)χΦ​(ρ) for the maximum ensemble Holevo information at mean ρ\rhoρ, H^Φ(ρ)\widehat H_\Phi(\rho)HΦ​(ρ) for the minimum average output entropy at that mean, and C‾(Φ;A)\overline C(\Phi;\mathcal A)C(Φ;A) for the maximum of χΦ\chi_\PhiχΦ​ over a nonempty closed mean constraint A\mathcal AA. The constraint A⊗B\mathcal A\otimes\mathcal BA⊗B includes every joint density with first marginal in A\mathcal AA and second marginal in B\mathcal BB, including correlated and entangled states.

Fix channels Φ\PhiΦ and Ψ\PsiΨ. The following six universally quantified properties are equivalent:

  1. For all nonempty closed A,B\mathcal A,\mathcal BA,B, C‾(Φ⊗Ψ;A⊗B)=C‾(Φ;A)+C‾(Ψ;B)\overline C(\Phi\otimes\Psi;\mathcal A\otimes\mathcal B)=\overline C(\Phi;\mathcal A)+\overline C(\Psi;\mathcal B)C(Φ⊗Ψ;A⊗B)=C(Φ;A)+C(Ψ;B).
  2. The same identity holds for all nonempty positive-observable linear inequality constraints with nonnegative budgets.
  3. For every joint state σ\sigmaσ, with σΦ=Tr⁡Kσ\sigma^\Phi=\operatorname{Tr}_K\sigmaσΦ=TrK​σ and σΨ=Tr⁡Hσ\sigma^\Psi=\operatorname{Tr}_H\sigmaσΨ=TrH​σ,
χΦ⊗Ψ(σ)≤χΦ(σΦ)+χΨ(σΨ).\chi_{\Phi\otimes\Psi}(\sigma)\le\chi_\Phi(\sigma^\Phi)+\chi_\Psi(\sigma^\Psi).χΦ⊗Ψ​(σ)≤χΦ​(σΦ)+χΨ​(σΨ).
  1. For every joint state, H^Φ⊗Ψ(σ)≥H^Φ(σΦ)+H^Ψ(σΨ)\widehat H_{\Phi\otimes\Psi}(\sigma)\ge\widehat H_\Phi(\sigma^\Phi)+\widehat H_\Psi(\sigma^\Psi)HΦ⊗Ψ​(σ)≥HΦ​(σΦ)+HΨ​(σΨ).
  2. For all states ρ,ω\rho,\omegaρ,ω, χΦ⊗Ψ(ρ⊗ω)=χΦ(ρ)+χΨ(ω)\chi_{\Phi\otimes\Psi}(\rho\otimes\omega)=\chi_\Phi(\rho)+\chi_\Psi(\omega)χΦ⊗Ψ​(ρ⊗ω)=χΦ​(ρ)+χΨ​(ω).
  3. For all states ρ,ω\rho,\omegaρ,ω, H^Φ⊗Ψ(ρ⊗ω)=H^Φ(ρ)+H^Ψ(ω)\widehat H_{\Phi\otimes\Psi}(\rho\otimes\omega)=\widehat H_\Phi(\rho)+\widehat H_\Psi(\omega)HΦ⊗Ψ​(ρ⊗ω)=HΦ​(ρ)+HΨ​(ω).

These are equivalences of properties of a fixed pair, not unconditional additivity claims.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Gap48_P46
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Gap48.P46

open scoped Matrix ComplexConjugate ENNReal ComplexOrder

open Filter

attribute [local instance] Classical.propDecidable

attribute [local instance] Classical.decEq

open QAlgorithms.Papers.Independent.Information100.P62 (tensorChannel)

open QAlgorithms.Papers.Independent.Information100.P54 (relativeEntropyBits)

open QAlgorithms.Papers.Independent.Dependency50.P23 (IsEB)

open QAlgorithms.Papers.Independent.Information100.P08
  (Ensemble ensembleAverage outputHolevo mixedRoof)

variable {I O J K : Type} [Fintype I] [Fintype O] [Fintype J] [Fintype K]

variable [DecidableEq I] [DecidableEq O] [DecidableEq J] [DecidableEq K]

/-- PDF pp.6–7, Theorem 1, including the product-state continuation: six equivalent
properties of the fixed pair. No unconditional additivity is asserted. -/
theorem additivity_equivalences [Nonempty I] [Nonempty O] [Nonempty J] [Nonempty K]
    (Phi : Superop I O) (Psi : Superop J K) (hPhi : IsChannel Phi) (hPsi : IsChannel Psi) :
    (AllConstraints Phi Psi ↔ LinearAdditive Phi Psi) ∧
    (AllConstraints Phi Psi ↔ ChiSubadditive Phi Psi) ∧
    (AllConstraints Phi Psi ↔ RoofSuperadditive Phi Psi) ∧
    (AllConstraints Phi Psi ↔ ProductChiAdditive Phi Psi) ∧
    (AllConstraints Phi Psi ↔ ProductRoofAdditive Phi Psi) := by sorry

end QAlgorithms.Papers.Independent.Gap48.P46

end
Source
On Shor's channel extension and constrained channels, arXiv:quant-ph/0306196v2, PDF p. 6, Theorem 1: four equivalent additivity properties
Read-back

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

Let I,O,J,KI,O,J,KI,O,J,K be finite nonempty types with decidable equality, and let Φ:I→O\Phi:I\to OΦ:I→O and Ψ:J→K\Psi:J\to KΨ:J→K denote maps between the corresponding complex matrix spaces. Assume each is a channel: it has a finite Kraus representation X↦∑jLjXLj†X\mapsto\sum_jL_jXL_j^\daggerX↦∑j​Lj​XLj†​, valid for every matrix XXX, with ∑jLj†Lj=1\sum_jL_j^\dagger L_j=1∑j​Lj†​Lj​=1. Write DV\mathcal D_VDV​ for the positive semidefinite trace-one matrices on VVV, and let Θ=Φ⊗Ψ\Theta=\Phi\otimes\PsiΘ=Φ⊗Ψ be the complex-linear map determined by Θ(Eii′⊗Ejj′)=Φ(Eii′)⊗Ψ(Ejj′)\Theta(E_{ii'}\otimes E_{jj'})=\Phi(E_{ii'})\otimes\Psi(E_{jj'})Θ(Eii′​⊗Ejj′​)=Φ(Eii′​)⊗Ψ(Ejj′​). For any such channel Λ\LambdaΛ, use all finite ensembles (pi,ρi)(p_i,\rho_i)(pi​,ρi​) with ρi∈DV\rho_i\in\mathcal D_Vρi​∈DV​, pi≥0p_i\ge0pi​≥0, and ∑ipi=1\sum_i p_i=1∑i​pi​=1, and define

χΛ(ρ)=sup⁡∑ipiρi=ρ[H(Λρ)−∑ipiH(Λρi)],H^Λ(ρ)=inf⁡∑ipiρi=ρ∑ipiH(Λρi),CΛ(A)=sup⁡ρ∈AχΛ(ρ),\begin{aligned} \chi_\Lambda(\rho)&=\sup_{\sum_i p_i\rho_i=\rho} \left[H(\Lambda\rho)-\sum_i p_iH(\Lambda\rho_i)\right],\\ \widehat H_\Lambda(\rho)&=\inf_{\sum_i p_i\rho_i=\rho}\sum_i p_iH(\Lambda\rho_i),\\ C_\Lambda(\mathcal A)&=\sup_{\rho\in\mathcal A}\chi_\Lambda(\rho), \end{aligned}χΛ​(ρ)HΛ​(ρ)CΛ​(A)​=∑i​pi​ρi​=ρsup​[H(Λρ)−i∑​pi​H(Λρi​)],=∑i​pi​ρi​=ρinf​i∑​pi​H(Λρi​),=ρ∈Asup​χΛ​(ρ),​

where H(X)=−∑jλjlog⁡2λjH(X)=-\sum_j\lambda_j\log_2\lambda_jH(X)=−∑j​λj​log2​λj​, with zero terms equal to 000; ensemble states need not be pure. For sets A⊆DI\mathcal A\subseteq\mathcal D_IA⊆DI​, B⊆DJ\mathcal B\subseteq\mathcal D_JB⊆DJ​, put M(A,B)={σ∈DI×J:Tr⁡Jσ∈A, Tr⁡Iσ∈B}\mathcal M(\mathcal A,\mathcal B)=\{\sigma\in\mathcal D_{I\times J}:\operatorname{Tr}_J\sigma\in\mathcal A,\ \operatorname{Tr}_I\sigma\in\mathcal B\}M(A,B)={σ∈DI×J​:TrJ​σ∈A, TrI​σ∈B}. Let A\mathsf AA be the assertion

CΘ(M(A,B))=CΦ(A)+CΨ(B)C_\Theta(\mathcal M(\mathcal A,\mathcal B)) =C_\Phi(\mathcal A)+C_\Psi(\mathcal B)CΘ​(M(A,B))=CΦ​(A)+CΨ​(B)

for every nonempty closed A⊆DI\mathcal A\subseteq\mathcal D_IA⊆DI​ and B⊆DJ\mathcal B\subseteq\mathcal D_JB⊆DJ​, without a convexity requirement. The theorem asserts that each of the following five assertions is equivalent to A\mathsf AA: the same capacity equality holds whenever A={ρ∈DI:Re⁡tr⁡(Aρ)≤α}\mathcal A=\{\rho\in\mathcal D_I:\operatorname{Re}\operatorname{tr}(A\rho)\le\alpha\}A={ρ∈DI​:Retr(Aρ)≤α} and B={ω∈DJ:Re⁡tr⁡(Bω)≤β}\mathcal B=\{\omega\in\mathcal D_J:\operatorname{Re}\operatorname{tr}(B\omega)\le\beta\}B={ω∈DJ​:Retr(Bω)≤β}, for all positive semidefinite A,BA,BA,B and all nonnegative α,β\alpha,\betaα,β making these sets nonempty; for every σ∈DI×J\sigma\in\mathcal D_{I\times J}σ∈DI×J​,

χΘ(σ)≤χΦ(Tr⁡Jσ)+χΨ(Tr⁡Iσ);\chi_\Theta(\sigma)\le\chi_\Phi(\operatorname{Tr}_J\sigma)+\chi_\Psi(\operatorname{Tr}_I\sigma);χΘ​(σ)≤χΦ​(TrJ​σ)+χΨ​(TrI​σ);

for every such σ\sigmaσ,

H^Φ(Tr⁡Jσ)+H^Ψ(Tr⁡Iσ)≤H^Θ(σ);\widehat H_\Phi(\operatorname{Tr}_J\sigma)+\widehat H_\Psi(\operatorname{Tr}_I\sigma) \le\widehat H_\Theta(\sigma);HΦ​(TrJ​σ)+HΨ​(TrI​σ)≤HΘ​(σ);

for every ρ∈DI,ω∈DJ\rho\in\mathcal D_I,\omega\in\mathcal D_Jρ∈DI​,ω∈DJ​,

χΘ(ρ⊗ω)=χΦ(ρ)+χΨ(ω);\chi_\Theta(\rho\otimes\omega)=\chi_\Phi(\rho)+\chi_\Psi(\omega);χΘ​(ρ⊗ω)=χΦ​(ρ)+χΨ​(ω);

and, for every such pair,

H^Θ(ρ⊗ω)=H^Φ(ρ)+H^Ψ(ω).\widehat H_\Theta(\rho\otimes\omega)=\widehat H_\Phi(\rho)+\widehat H_\Psi(\omega).HΘ​(ρ⊗ω)=HΦ​(ρ)+HΨ​(ω).
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