Theorem I.1: Dbar²≤1−Fbar≤Dbar; for quantum encoders also 1−Fbar≤1−Fe.
ProvedQAlgorithms.Papers.Independent.Renew46.P43.sourceQualityComparisonLet a finite pure-state ensemble on a nonzero finite-dimensional Hilbert space generate iid words, and let a positive-length compression code have arbitrary state-valued encoding and a CPTP decoder. Write bar F for mean word overlap and bar D for mean half trace norm. For a quantum encoder, F_e is the purification overlap of its CPTP extension.
This is the source result with its stated operational or information-theoretic domain.
import Definitions.Def_QAlgorithms_Papers_Independent_Renew46_P43
noncomputable section
namespace QAlgorithms.Papers.Independent.Renew46.P43
open scoped BigOperators Matrix ComplexConjugate ComplexOrder ENNReal Matrix.Norms.L2Operator
open Filter MeasureTheory
open QAlgorithms.Papers.Independent.Information100.P08 (IsPure)
open QAlgorithms.Papers.Independent.Information100.P06 (Below)
open QAlgorithms.Papers.Independent.Information100.P60 (copies relativeBits)
open QAlgorithms.Papers.Independent.Dependency50.P19 (count both)
open QAlgorithms.Papers.Independent.Information100.P22 (coherentInformation)
attribute [local instance] Classical.propDecidable Classical.decEq
/-- Theorem I.1, PDF p.8. The first chain holds for every arbitrary encoder;
the entanglement-fidelity clause uses each actual quantum extension separately. -/
theorem sourceQualityComparison {a d n : ℕ} {S : PureSource a d} (c : SourceCode S n) :
c.distortion^2 ≤ 1 - c.fidelity ∧ 1 - c.fidelity ≤ c.distortion ∧
∀ E, EncoderExtension c E →
1 - c.fidelity ≤ 1 - sourceFe (copies n S.average) (c.decode ∘ E) := by sorry
end QAlgorithms.Papers.Independent.Renew46.P43
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be natural numbers, and let be a source with distinct pure states on , where , together with a probability mass function on its labels; probabilities may be zero. Let be a code of length , with a compressed dimension , an arbitrary assignment of density matrices on to all words , and a decoder channel into . Here a density matrix is positive semidefinite with trace one, and a channel has a finite Kraus representation with . Put , , and . Define the code’s average overlap and average trace distance by
where . Then . Moreover, for every channel from the word space to satisfying for every word, including words of zero probability,
For a matrix and a map , the quantity used here is , where , the square root is defined by continuous functional calculus, and applies to each matrix block indexed by the first factor.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.