Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Section 3.1 / (15): bottleneck inequality

Open
QAlgorithms.Papers.Independent.Gap20.P07.capacity_bottleneck

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

prove2quantum

Let T2:H1→H2T_2:H_1\to H_2T2​:H1​→H2​ and T1:H2→H3T_1:H_2\to H_3T1​:H2​→H3​ be composable CPTP channels, and let SSS be any CPTP reference channel, all on nonzero complex Hilbert spaces. For the full-cb relative capacities of Definition 2.1,

Q(T1∘T2,S)≤min⁡{Q(T1,S),Q(T2,S)}.Q(T_1\circ T_2,S)\le\min\{Q(T_1,S),Q(T_2,S)\}.Q(T1​∘T2​,S)≤min{Q(T1​,S),Q(T2​,S)}.

The reference may be noisy. This is the capacity bottleneck property for sequential channel use.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Gap20_P07
Formal statement
noncomputable section

open Filter MeasureTheory

open scoped Topology ENNReal NNReal Matrix ComplexOrder ComplexConjugate

attribute [local instance] Classical.propDecidable

attribute [local instance] Classical.decEq

attribute [local instance] IsStarNormal.instContinuousFunctionalCalculus

namespace QAlgorithms.Papers.Independent.Gap20.P07

open QAlgorithms

open QAlgorithms.RealRegisters (Hilbert)

open QAlgorithms.Papers.Independent.Dependency50.P33
  (State IsTraceClass IsTraceClassChannel traceSize rankOne tensorApply)

open QAlgorithms.Papers.Independent.Frontier30.P14
  (Operation isometryCoeff conjugateKernel)

open QAlgorithms.Papers.Independent.Claimed50.P35
  (stabilizedTraceClassNorm traceClassTensorMap traceClassChannelPower
   transpose transposeBound TraceLinear tracePair MeasurePrepare)

open QAlgorithms.Papers.Independent.Gap48.P29 (fe FiniteCPMap)

/-- PDF p.8, (15): bottleneck inequality, including arbitrary noisy references. -/
theorem capacity_bottleneck {I J K A B : Type}
    [Nonempty I] [Nonempty J] [Nonempty K] [Nonempty A] [Nonempty B]
    (T1 : Operation J K) (T2 : Operation I J) (S : Operation A B)
    (h1 : IsTraceClassChannel T1) (h2 : IsTraceClassChannel T2) (hS : IsTraceClassChannel S) :
    relativeCapacity (T1 ∘ T2) S ≤ min (relativeCapacity T1 S) (relativeCapacity T2 S) := by sorry

end QAlgorithms.Papers.Independent.Gap20.P07

end
Source
Tema Con Variazioni: Quantum Channel Capacity, arXiv:quant-ph/0311037v1, PDF p. 8, Section 3.1 / (15): bottleneck inequality
Read-back

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

Let I,J,K,A,BI,J,K,A,BI,J,K,A,B be nonempty types, and let T1:CJ×J→CK×KT_1:\mathbb C^{J\times J}\to\mathbb C^{K\times K}T1​:CJ×J→CK×K, T2:CI×I→CJ×JT_2:\mathbb C^{I\times I}\to\mathbb C^{J\times J}T2​:CI×I→CJ×J, and S:CA×A→CB×BS:\mathbb C^{A\times A}\to\mathbb C^{B\times B}S:CA×A→CB×B be channels in the following sense. A matrix XXX is trace class when Xij=∑r≥0ur(i)vr(j)‾X_{ij}=\sum_{r\ge0}u_r(i)\overline{v_r(j)}Xij​=∑r≥0​ur​(i)vr​(j)​ for square-summable vectors with ∑r∥ur∥∥vr∥<∞\sum_r\|u_r\|\|v_r\|<\infty∑r​∥ur​∥∥vr​∥<∞; τ(X)\tau(X)τ(X) is the infimum of these sums in [0,∞][0,\infty][0,∞], including +∞+\infty+∞ if there is no expansion. A density is trace class, has positive semidefinite finite principal restrictions, and has summable real diagonal of sum 111. A channel preserves trace-class matrices, is additive and complex-homogeneous on them, and preserves densities after adjoining any finite ancillary register. For any matrix map FFF, let ∥F∥st\|F\|_{\mathrm{st}}∥F∥st​ be the supremum of τ((F⊗idr)X)\tau((F\otimes\mathrm{id}_r)X)τ((F⊗idr​)X) over integers r≥0r\ge0r≥0 and trace-class XXX with τ(X)≤1\tau(X)\le1τ(X)≤1, where

((F⊗idr)X)(j,a),(j′,b)=F((X(i,a),(i′,b))i,i′)j,j′.((F\otimes\mathrm{id}_r)X)_{(j,a),(j',b)} =F((X_{(i,a),(i',b)})_{i,i'})_{j,j'}.((F⊗idr​)X)(j,a),(j′,b)​=F((X(i,a),(i′,b)​)i,i′​)j,j′​.

For maps PPP and QQQ, define ΔP,Q(n,m)=inf⁡E,D∥D∘P[n]∘E−Q[m]∥st\Delta_{P,Q}(n,m)=\inf_{E,D}\|D\circ P^{[n]}\circ E-Q^{[m]}\|_{\mathrm{st}}ΔP,Q​(n,m)=infE,D​∥D∘P[n]∘E−Q[m]∥st​, where n,mn,mn,m are positive integers and E,DE,DE,D are channels between the corresponding input and output registers. Here P[n]P^{[n]}P[n] is formed by applying PPP to the first-coordinate matrix block and recursively to the remaining coordinates; its zeroth power evaluates at the unique empty tuples. Define C(P,Q)C(P,Q)C(P,Q) as the supremum, in [0,∞][0,\infty][0,∞], of all real R>0R>0R>0 for which every pair of positive-integer sequences na,man_a,m_ana​,ma​, satisfying na→∞n_a\to\inftyna​→∞ and lim sup⁡ama/na≤R\limsup_a m_a/n_a\le Rlimsupa​ma​/na​≤R, satisfies ΔP,Q(na,ma)→0\Delta_{P,Q}(n_a,m_a)\to0ΔP,Q​(na​,ma​)→0. An empty rate supremum is 000; no attainment of the error infimum is required. Then

C(T1∘T2,S)≤min⁡{C(T1,S),C(T2,S)}.C(T_1\circ T_2,S)\le \min\{C(T_1,S),C(T_2,S)\}.C(T1​∘T2​,S)≤min{C(T1​,S),C(T2​,S)}.
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