Theorem 18: entangled ensemble trade-off
OpenQAlgorithms.Papers.Independent.Intakenext100.P54.entangled_ensemble_tradeoffLet be a finite normalized pure bipartite ensemble in fixed positive dimensions, and let . Define its operational ebit curve using one-way input-dependent Alice CP instruments, message-dependent arbitrary CPTP Bob decoders, iid product blocks, global average fidelity, and both positive rate slacks and errors before every sufficiently large blocklength. Then
The classical auxiliary is finite but arbitrary. At feasible rates the minimum is attained and its rate pair is achievable; an infeasible rate has value . The cost averages receiver conditional entropies. In particular it is not replaced by . Although the operational target preserves Alice's purification, the information curve depends only on the receiver marginal ensemble.
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)
/-- Theorem 18, PDF pp.10–11: minimise S(B|C), not I(X;B|C), over all finite stochastic
auxiliaries subject to I(X;BC) ≤ R. The operational target includes Alice's purification. -/
theorem entangled_ensemble_tradeoff {m a b : ℕ} (ha : 0 < a) (hb : 0 < b)
(E : BipartiteEnsemble m a b) (R : ℝ) (hR : 0 ≤ R) :
bipOperational E R = rspSingleLetter E.prob (bobStates E) R ∧
rspMinimumAttained E.prob (bobStates E) R ∧
(bipOperational E R ≠ ⊤ → BipartiteAchievable E R (bipOperational E R).toReal) := 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 , ensemble of probabilities summing to and unit vectors , and real , put and let denote the specified continuous-functional-calculus entropy. Define an operationally feasible number to mean and that, for every , every sufficiently large positive integer admits positive integers , a natural number , matrices with , and finite Kraus maps with , whose outputs
satisfy , , and , where . Let be the infimum in of these feasible . For an auxiliary natural number and numbers with for every , set and
Scalar division by zero is , so the vector inside is zero when . Let in , over all such . The theorem asserts ; if , some auxiliary satisfies and ; and if , the real value of is operationally feasible. Empty feasible sets have infimum , and the ensemble requirements exclude .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.