Section 7.4: ensemble–POVM duality at a faithful average
OpenQAlgorithms.Papers.Independent.Intakethird100.P51.ensemblePOVMDualityFix a faithful normalized state . From any finite ensemble with and average , define
These effects form a POVM. Conversely, for a POVM of nonzero effects, put
The sum to one, every is a density matrix, and . Both transformations are inverse, including recovery of the original weights and component states. This gives the normalized ensemble–POVM duality after deleting null entries.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakethird100_P51
noncomputable section
open scoped Matrix ComplexOrder ComplexConjugate Matrix.Norms.L2Operator
open MeasureTheory
namespace QAlgorithms.Papers.Independent.Intakethird100.P51
attribute [local instance] Classical.propDecidable Classical.decEq
-- Exact interfaces from the frozen accepted library; no local redeclarations.
open QAlgorithms.Papers.Independent.Information100.P63 (Mat)
open QAlgorithms.Papers.Independent.Information100.P56 (entropy holevo IsLaw)
open QAlgorithms.Papers.Independent.Information100.P27 (average)
open QAlgorithms.Papers.Independent.Dependency50.P21 (relativeEntropy marginalAB channelAdjoint)
open QAlgorithms.Papers.Independent.Dependency50.P35 (marginalAC)
open QAlgorithms.Papers.Independent.Gap48.P19 (shannonNats)
open QAlgorithms.Papers.Independent.Gap48.P33 (measurementInformation)
section Statements
variable {A B C D J K O : Type}
variable [Fintype A] [Fintype B] [Fintype C] [Fintype D] [Fintype J] [Fintype K] [Fintype O]
variable [DecidableEq A] [DecidableEq B] [DecidableEq C] [DecidableEq D]
variable [DecidableEq J] [DecidableEq K] [DecidableEq O]
/-- Section 7.4, PDF p.21: positivity, normalization and both inverse properties
of the faithful-reference ensemble–POVM duality, after deleting null entries. -/
theorem ensemblePOVMDuality [Nonempty A] [Nonempty J] [Nonempty O] (rho : Mat A) (hr : FaithfulDensity rho) :
(∀ w : J → ℝ, (∀ j, 0 < w j) → (∑ j, w j) = 1 →
∀ sigma : J → Mat A, (∀ j, IsDensityMatrix (sigma j)) → average w sigma = rho →
IsPOVM (dualEffects rho w sigma) ∧
(∀ j, effectProb (dualEffects rho w sigma j) rho = w j) ∧
(∀ j, dualStates rho (dualEffects rho w sigma) j = sigma j)) ∧
(∀ E : O → Mat A, IsPOVM E → (∀ b, E b ≠ 0) →
(∀ b, 0 < effectProb (E b) rho) ∧
IsLaw (fun b => effectProb (E b) rho) ∧
(∀ b, IsDensityMatrix (dualStates rho E b)) ∧
average (fun b => effectProb (E b) rho) (dualStates rho E) = rho ∧
dualEffects rho (fun b => effectProb (E b) rho) (dualStates rho E) = E) := by sorry
end Statements
end QAlgorithms.Papers.Independent.Intakethird100.P51
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For finite nonempty index types , with decidable equality as required for the matrix definitions, let be a positive definite matrix on with trace one, and define and by continuous functional calculus. First, for every strictly positive family summing to one and every family of positive semidefinite trace-one matrices satisfying , the matrices are positive semidefinite and sum to , satisfy , and satisfy
Second, for every family of positive semidefinite matrices summing to , with for every , the numbers are strictly positive and sum to one; every is positive semidefinite of trace one; ; and the family equals . Thus all denominators in these asserted constructions are positive under their respective hypotheses.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.