Fixed-reference finite-block AEP
ProvedQAlgorithms.Papers.Independent.Support50.P23.fixedReferenceFiniteBlockLet and be normalized finite-dimensional states and let . Define and . On the branch , if and , then
If the support of is not contained in that of , the fixed-reference von Neumann entropy is , yielding the trivial lower bound. For , the smooth fixed min entropy is . This version retains the exact reference tensor power.
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P23
noncomputable section
open Filter
open scoped ENNReal ComplexConjugate ComplexOrder
namespace QAlgorithms.Papers.Independent.Support50.P23
attribute [local instance] Classical.propDecidable
-- Exact accepted bodies. Other similarly named catalogue entries use different
-- smoothing, conditioning order, logarithm base, or channel-level objects.
open QAlgorithms.Papers.Independent.Gap20.P03 (fixedMin)
open QAlgorithms.Papers.Independent.Dependency50.P36 (relativeEntropy)
open QAlgorithms.Papers.Independent.Information100.P67 (IsSubnormalized)
open QAlgorithms.Papers.Independent.Information100.P69 (matrixPower)
variable {A B C D I : Type*}
variable [Fintype A] [Fintype B] [Fintype C] [Fintype D] [Fintype I]
variable [DecidableEq A] [DecidableEq B] [DecidableEq C] [DecidableEq D] [DecidableEq I]
/-- Remark 10, PDF p.6: sigma^{tensor n} remains fixed. Use the finite-eta formula
only on its real domain; support failure is a -infinity lower bound. The epsilon
domain correction is the same as in Theorem 9. -/
theorem fixedReferenceFiniteBlock [Nonempty A] [Nonempty B]
(rho : Matrix (A × B) (A × B) ℂ) (sigma : Matrix B B ℂ)
(hrho : IsDensityMatrix rho) (hsigma : IsDensityMatrix sigma)
(epsilon : ℝ) (hepsilon : 0 < epsilon) (n : ℕ) (hn : 0 < n) :
((epsilon < 1 ∧ entropyConvergenceParameter rho sigma ≠ ⊤ ∧
(8 / 5 : ℝ) * Real.logb 2 (2 / epsilon ^ 2) ≤ (n : ℝ)) →
fixedVonNeumann rho sigma -
((aepBlockError epsilon (entropyConvergenceParameter rho sigma) / Real.sqrt (n : ℝ) : ℝ) : EReal) ≤
(((n : ℝ)⁻¹ : ℝ) : EReal) *
sineSmoothFixedMin (jointPower rho n) (matrixPower sigma n) epsilon) ∧
(¬ supportedBy rho (identityTensorReference (A := A) sigma) →
fixedVonNeumann rho sigma = ⊥ ∧ fixedVonNeumann rho sigma ≤
(((n : ℝ)⁻¹ : ℝ) : EReal) *
sineSmoothFixedMin (jointPower rho n) (matrixPower sigma n) epsilon) ∧
(1 ≤ epsilon → sineSmoothFixedMin (jointPower rho n) (matrixPower sigma n) epsilon = ⊤) := by sorry
end QAlgorithms.Papers.Independent.Support50.P23
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be nonempty finite types with decidable equality, and let on and on be positive semidefinite matrices of trace . Write , and say that is supported by when . Put , using functional calculus with the scalar function , and define when this support inclusion holds, and otherwise. For , define to be at , to be when and the support inclusion fails, and otherwise to be ; here applies to positive eigenvalues and assigns at zero, while and for . Let , , and for finite , and put and , where is the real conversion of the extended nonnegative number , assigning to . Define
with the suprema taken in the extended reals. For every real and natural , all three assertions hold: if , , and , then ; if , then and ; and if , including , then . The tensor powers have entries given by products over the coordinates, and the smoothing permits and does not require trace .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.