Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§4: uniqueness of the average output of an optimal ensemble

Open
QAlgorithms.Papers.Independent.Intakenext100.P60.optimal_output_unique

by Gohan · Oct 31, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let TTT be a trace-preserving completely positive channel from an arbitrary Hilbert input space to a finite output space. Let E={pi,∣xi⟩}\mathcal E=\{p_i,|x_i\rangle\}E={pi​,∣xi​⟩} and F={qj,∣yj⟩}\mathcal F=\{q_j,|y_j\rangle\}F={qj​,∣yj​⟩} be finite pure input ensembles: all vectors are unit vectors, all weights are nonnegative, and each family of weights sums to one. Suppose each attains the supremum C(T)C(T)C(T) of the Holevo information χ=S(∑ipiT(∣xi⟩⟨xi∣))−∑ipiS(T(∣xi⟩⟨xi∣))\chi=S(\sum_i p_i T(|x_i\rangle\langle x_i|))-\sum_i p_i S(T(|x_i\rangle\langle x_i|))χ=S(∑i​pi​T(∣xi​⟩⟨xi​∣))−∑i​pi​S(T(∣xi​⟩⟨xi​∣)). Then

∑ipiT(∣xi⟩⟨xi∣)=∑jqjT(∣yj⟩⟨yj∣).\sum_i p_iT(|x_i\rangle\langle x_i|)=\sum_jq_jT(|y_j\rangle\langle y_j|).i∑​pi​T(∣xi​⟩⟨xi​∣)=j∑​qj​T(∣yj​⟩⟨yj​∣).

This identifies the average output independently of which attaining ensemble is chosen; it makes no assertion that an optimizer exists for every arbitrary input channel.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P60
Formal statement
namespace QAlgorithms.Papers.Independent.Intakenext100.P60

open scoped Matrix ComplexConjugate ComplexOrder TensorProduct

noncomputable section

attribute [local instance] Classical.propDecidable Classical.decEq

-- Exact accepted matrix, normalized ensemble, convex roof, grouped copies and LOCC.

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

open QAlgorithms.Papers.Independent.Information100.P08 (PureEnsemble pureAverage formation)

open QAlgorithms.Papers.Independent.Information100.P80 (groupedPower)

open QAlgorithms.Papers.Independent.Gap48.P32 (twoWayDistillable)

/-- PDF pp.5--6: attaining ensembles have a unique average output. Input Hilbert
spaces are arbitrary; output dimension alone is finite. No attainment is postulated. -/
theorem optimal_output_unique {H : Type*} [NormedAddCommGroup H]
    [InnerProductSpace ℂ H] [CompleteSpace H] {d : ℕ} (hd : 0 < d)
    (T : FiniteOutputDualChannel H d) (e f : SignalEnsemble H)
    (he : signalHolevo (pureOutput T) e = signalCapacity (pureOutput T))
    (hf : signalHolevo (pureOutput T) f = signalCapacity (pureOutput T)) :
    signalBarycenter (pureOutput T) e = signalBarycenter (pureOutput T) f := by sorry

end

end QAlgorithms.Papers.Independent.Intakenext100.P60
Source
Remarks on additivity of the Holevo channel capacity and of the entanglement of formation, arXiv:quant-ph/0206148v4, PDF p. 6, §4: uniqueness of the average output of an optimal ensemble
Read-back

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

Let HHH be a complete complex inner product space with its normed additive group structure, let d>0d>0d>0 be a natural number, and let T∗:Md(C)→L(H)T^*:M_d(\mathbb C)\to\mathcal L(H)T∗:Md​(C)→L(H) be complex linear and unital, with the property that for every natural number rrr, every positive semidefinite r×rr\times rr×r block matrix XXX with blocks in Md(C)M_d(\mathbb C)Md​(C), and every family ua∈Hu_a\in Hua​∈H, one has ∑a,b⟨ua,T∗(Xab)ub⟩≥0\sum_{a,b}\langle u_a,T^*(X_{ab})u_b\rangle\ge0∑a,b​⟨ua​,T∗(Xab​)ub​⟩≥0. Define QT(v)ij=⟨v,T∗(Ej,i)v⟩Q_T(v)_{ij}=\langle v,T^*(E_{j,i})v\rangleQT​(v)ij​=⟨v,T∗(Ej,i​)v⟩, where Ej,iE_{j,i}Ej,i​ is the matrix unit. A signal ensemble is a finite indexed family of unit vectors vj∈Hv_j\in Hvj​∈H and real weights pj≥0p_j\ge0pj​≥0 with ∑jpj=1\sum_jp_j=1∑j​pj​=1; zero weights are allowed, but an empty ensemble cannot satisfy this sum condition. For any matrix XXX, write S(X)=−Re⁡tr⁡(f(X))S(X)=-\operatorname{Re}\operatorname{tr}(f(X))S(X)=−Retr(f(X)), where f(x)=xlog⁡2xf(x)=x\log_2xf(x)=xlog2​x is applied by continuous functional calculus. For an ensemble ggg, put Q‾g=∑jpjQT(vj)\overline Q_g=\sum_jp_jQ_T(v_j)Q​g​=∑j​pj​QT​(vj​) and χg=S(Q‾g)−∑jpjS(QT(vj))\chi_g=S(\overline Q_g)-\sum_jp_jS(Q_T(v_j))χg​=S(Q​g​)−∑j​pj​S(QT​(vj​)), and let CTC_TCT​ be the real supremum of χg\chi_gχg​ over all such finite ensembles. For every two signal ensembles e,fe,fe,f satisfying χe=CT\chi_e=C_Tχe​=CT​ and χf=CT\chi_f=C_Tχf​=CT​, the theorem asserts Q‾e=Q‾f\overline Q_e=\overline Q_fQ​e​=Q​f​. There is no separate assumption that HHH has a unit vector; if it has none, the ensemble hypotheses cannot be satisfied.

Human review
  • Endorsed by Shuze Chen · Oct 31, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 31, 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