Section VI.C: cq quantum reverse Shannon application
OpenQAlgorithms.Papers.Independent.Intakenext100.P54.cq_reverse_shannon_applicationFor a fixed finite ensemble of normalized pure bipartite states, let be the receiver marginals and . For every positive error and rate slack, all sufficiently large blocklengths admit one code simultaneously approximating the chosen product purifications in global average overlap and their receiver marginals in average squared Uhlmann fidelity, with
Both average fidelities are at least one minus the requested error, and both support rates are bounded by the displayed rates plus the requested slack. Alice's share is discarded only when evaluating the second fidelity. This gives the stated cq reverse Shannon application for the fixed ensemble law, without a channel diamond-norm assertion.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P54
noncomputable section
open scoped BigOperators Matrix ComplexConjugate ComplexOrder NNReal ENNReal
open MeasureTheory ProbabilityTheory
namespace QAlgorithms.Papers.Independent.Intakenext100.P54
open QAlgorithms.Papers.Independent.Information100.P03 (PureEnsemble)
open QAlgorithms.Papers.Independent.Information100.P53 (fidelity)
open QAlgorithms.Papers.Independent.Dependency50.P07 (wordProb)
open QAlgorithms.Papers.Independent.Claimed50.P08 (classicalEntropy)
open QAlgorithms.Papers.Independent.Claimed50.P38 (tensorSuperop)
/-- Section VI.C, PDF p.13: the preparation scheme simultaneously approximates the chosen
purifications and the receiver's mixed-state ensemble at Holevo cbit rate. This is a fixed-law
average-fidelity assertion, not a diamond-norm simulation claim. -/
theorem cq_reverse_shannon_application {m a b : ℕ} (ha : 0 < a) (hb : 0 < b)
(E : BipartiteEnsemble m a b) :
∀ eps : ℝ, 0 < eps → ∀ delta : ℝ, 0 < delta →
∃ nzero : ℕ, 0 < nzero ∧ ∀ n ≥ nzero,
∃ P : BipartiteCode (Fin n → Fin m) (Fin n → Fin a) (Fin n → Fin b),
1 - eps ≤ bipAverage E n P ∧ 1 - eps ≤ marginalAverageFidelity E n P ∧
Real.logb 2 P.K / n ≤ bobHolevo E + delta ∧
Real.logb 2 P.S / n ≤ bobAverageEntropy E + delta := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P54
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every natural number , positive natural numbers , and ensemble of unit vectors with probabilities summing to , put , , , and , where for , evaluated by continuous functional calculus. For every and , there exists a positive natural number such that every natural number admits a code with positive integers , a natural number , matrices satisfying for each word , and channels to . Each channel has a finite Kraus representation with the sum of the Kraus adjoint-products equal to the identity. With , , and output
this code satisfies all four bounds
where . There is no upper restriction on or ; the ensemble conditions exclude .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.