Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§7.3: input concavity of coherent information

Proved
QAlgorithms.Papers.Independent.Support50.P02.degradable_concavity

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

prove2quantum

Let U:A′→BEU:A'\to BEU:A′→BE be an isometry of finite, positive-dimensional complex registers. Write N(ρ)=Tr⁡E(UρU†)\mathcal N(\rho)=\operatorname{Tr}_E(U\rho U^\dagger)N(ρ)=TrE​(UρU†) and Nc(ρ)=Tr⁡B(UρU†)\mathcal N^c(\rho)=\operatorname{Tr}_B(U\rho U^\dagger)Nc(ρ)=TrB​(UρU†) for its receiver channel and complement. Suppose a CPTP map D:B→E\mathcal D:B\to ED:B→E satisfies Nc=D∘N\mathcal N^c=\mathcal D\circ\mathcal NNc=D∘N. Define Ic(ρ,N)=H(N(ρ))−H(Nc(ρ))I_c(\rho,\mathcal N)=H(\mathcal N(\rho))-H(\mathcal N^c(\rho))Ic​(ρ,N)=H(N(ρ))−H(Nc(ρ)), with entropy in bits. For density operators ρ0,ρ1\rho_0,\rho_1ρ0​,ρ1​ and 0≤λ≤10\le\lambda\le10≤λ≤1,

λIc(ρ0,N)+(1−λ)Ic(ρ1,N)≤Ic(λρ0+(1−λ)ρ1,N).\lambda I_c(\rho_0,\mathcal N)+(1-\lambda)I_c(\rho_1,\mathcal N)\le I_c(\lambda\rho_0+(1-\lambda)\rho_1,\mathcal N).λIc​(ρ0​,N)+(1−λ)Ic​(ρ1​,N)≤Ic​(λρ0​+(1−λ)ρ1​,N).

This is concavity in the input state of a fixed degradable channel, including both endpoints.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P02
Formal statement
noncomputable section

open scoped BigOperators Matrix ComplexConjugate

namespace QAlgorithms.Papers.Independent.Support50.P02

open QAlgorithms.Papers.Independent.Information100.P67
  (squaredFidelity receiverChannel complementaryChannel IsDegradable)

open QAlgorithms.Papers.Independent.Information100.P21 (Op)

open QAlgorithms.Papers.Independent.Information100.P03 (PureEnsemble)

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

attribute [local instance] Classical.propDecidable Classical.decEq

variable {A B C E R S I J : Type*}

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

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

variable [Fintype R] [DecidableEq R] [Fintype S] [DecidableEq S]

variable [Fintype I] [DecidableEq I] [Fintype J] [DecidableEq J]

/-- Concavity in the INPUT of a degradable channel, §7.3, PDF p.35.
This is distinct from background convexity in the channel at fixed input (p.4). -/
theorem degradable_concavity [Nonempty A] [Nonempty B] [Nonempty E]
    (U : Matrix (B × E) A ℂ) (hU : IsDegradable U)
    (rho0 rho1 : Op A) (h0 : IsDensityMatrix rho0) (h1 : IsDensityMatrix rho1)
    (lambda : ℝ) (hl0 : 0 ≤ lambda) (hl1 : lambda ≤ 1) :
    lambda * inputCoherent U rho0 + (1 - lambda) * inputCoherent U rho1 ≤
      inputCoherent U ((lambda : ℂ) • rho0 + ((1 - lambda : ℝ) : ℂ) • rho1) := by sorry

end QAlgorithms.Papers.Independent.Support50.P02

end
Source
Capacity Theorems for Quantum Multiple Access Channels: Classical-Quantum and Quantum-Quantum Capacity Regions, arXiv:quant-ph/0501045v3, PDF p. 35, §7.3: input concavity of coherent information
Read-back

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

For arbitrary nonempty finite types A,B,EA,B,EA,B,E, each with decidable equality, let UUU be a complex matrix from AAA to B×EB\times EB×E satisfying U†U=IU^\dagger U=IU†U=I. Suppose there is a map DDD from BBB-indexed matrices to EEE-indexed matrices with a finite Kraus representation D(X)=∑jKjXKj†D(X)=\sum_jK_jXK_j^\daggerD(X)=∑j​Kj​XKj†​, ∑jKj†Kj=I\sum_jK_j^\dagger K_j=I∑j​Kj†​Kj​=I, such that for every AAA-indexed complex matrix XXX,

tr⁡B(UXU†)=D ⁣(tr⁡E(UXU†)).\operatorname{tr}_B(UXU^\dagger) =D\!\left(\operatorname{tr}_E(UXU^\dagger)\right).trB​(UXU†)=D(trE​(UXU†)).

For every pair of positive semidefinite AAA-indexed matrices ρ0,ρ1\rho_0,\rho_1ρ0​,ρ1​ with complex trace equal to 111, and every real λ∈[0,1]\lambda\in[0,1]λ∈[0,1], the assertion is

λIU(ρ0)+(1−λ)IU(ρ1)≤IU ⁣(λρ0+(1−λ)ρ1),\lambda I_U(\rho_0)+(1-\lambda)I_U(\rho_1) \le I_U\!\left(\lambda\rho_0+(1-\lambda)\rho_1\right),λIU​(ρ0​)+(1−λ)IU​(ρ1​)≤IU​(λρ0​+(1−λ)ρ1​),

where IU(X)=S(tr⁡EUXU†)−S(tr⁡BUXU†)I_U(X)=S(\operatorname{tr}_E UXU^\dagger)-S(\operatorname{tr}_B UXU^\dagger)IU​(X)=S(trE​UXU†)−S(trB​UXU†) and S(Y)=−Re⁡tr⁡[f(Y)]S(Y)=-\operatorname{Re}\operatorname{tr}[f(Y)]S(Y)=−Retr[f(Y)] for the continuous functional calculus of f(t)=tlog⁡2tf(t)=t\log_2tf(t)=tlog2​t. The partial traces sum the diagonal indices of the indicated subsystem. The endpoints λ=0,1\lambda=0,1λ=0,1 and the case ρ0=ρ1\rho_0=\rho_1ρ0​=ρ1​ are included.

Human review
  • Endorsed by Shuze Chen · Oct 23, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Goku · Oct 23, 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