§6.2, equations (26)–(28): optimized entanglement fidelity
ProvedQAlgorithms.Papers.Independent.Support50.P41.processedFanoFor a density matrix , CPTP map on dimension , and normalized purification , the maximum over system unitaries is attained. Its value is independent of the purification, and
The logarithmic bound is for ; at , and . These bounds retain the original channel's entropy because the processing is unitary.
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P41
noncomputable section
namespace QAlgorithms.Papers.Independent.Support50.P41
open QAlgorithms
open QAlgorithms.Entanglement (IsCompletelyPositive)
open QAlgorithms.Papers.Independent.Dependency50.P40 (IsPurification purificationFidelity)
open QAlgorithms.Papers.Independent.Dependency50.P45 (entropyExchange)
open QAlgorithms.Papers.Independent.Information100.P08 (canonicalPurification)
open QAlgorithms.Papers.Independent.Information100.P34 (choi dephasing)
open QAlgorithms.Papers.Independent.Dependency50.P38 (vectorizeInputFirst)
open QAlgorithms.Papers.Independent.Gap20.P06 (dilationChannel environmentOutput dilatedVector)
open QAlgorithms.Papers.Independent.Information100.P50 (complementaryMap)
open QAlgorithms.Papers.Independent.Renew46.P43 (IsCountablePOVM classicalKLTerm countableEntropy)
open scoped Matrix ComplexConjugate ComplexOrder ENNReal
attribute [local instance] Classical.propDecidable Classical.decEq
-- The accepted generic dilation functions fix classical equality on each factor.
-- Match that instance for Fin before forming their product unitary groups.
local instance (priority := 2000) (n : ℕ) : DecidableEq (Fin n) := Classical.decEq _
/-- PDF pp.27--28, (26)--(28): attained unitary-processing maximum and Fano with original entropy. -/
theorem processedFano {d a : ℕ} [NeZero d]
(Phi : Superop (Fin d) (Fin d)) (hPhi : IsChannel Phi)
(rho : Matrix (Fin d) (Fin d) ℂ) (hrho : IsDensityMatrix rho)
(psi : EuclideanSpace ℂ (Fin a × Fin d)) (hpsi : IsPurification rho psi) :
∃ U : Matrix.unitaryGroup (Fin d) ℂ,
let f := purificationFidelity psi (conjSuperop (U : Matrix (Fin d) (Fin d) ℂ) ∘ Phi)
(∀ V : Matrix.unitaryGroup (Fin d) ℂ,
purificationFidelity psi (conjSuperop (V : Matrix (Fin d) (Fin d) ℂ) ∘ Phi) ≤ f) ∧
purificationFidelity psi Phi ≤ f ∧
(∀ b (phi : EuclideanSpace ℂ (Fin b × Fin d)), IsPurification rho phi →
∀ V : Matrix.unitaryGroup (Fin d) ℂ,
purificationFidelity phi (conjSuperop (V : Matrix (Fin d) (Fin d) ℂ) ∘ Phi) =
purificationFidelity psi (conjSuperop (V : Matrix (Fin d) (Fin d) ℂ) ∘ Phi)) ∧
(d = 1 → f = 1 ∧ entropyExchange Phi rho = 0) ∧
(2 ≤ d → entropyExchange Phi rho ≤ binEntropy2 f + (1-f)*Real.logb 2 ((d : ℝ)^2-1)) ∧
entropyExchange Phi rho ≤ 1+2*(1-f)*Real.logb 2 d := by sorry
end QAlgorithms.Papers.Independent.Support50.P41
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let and , let admit a finite representation with , let have trace , and let have norm and partial trace . Write , , and , where ; matrix functions use continuous functional calculus and zero logarithmic terms are zero. There exists a unitary such that, with , every unitary satisfies , and ; moreover, for every natural number , every norm-one with , and every such , one has . The same satisfies
where . Zero reference dimensions and are included in the quantifiers, but cannot support the required norm-one purifications.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.