Range and sign of the relative advantage parameter
ProvedQAlgorithms.Papers.Independent.Bridge50.P20.advantage_criterionprove2quantum
Let and denote, respectively, a finite nonnegative squashed-entanglement cut cost and a finite nonnegative versatile cut cost, with . Define . Then
This compares the cut-cost terms used in the small-error regime. It does not assert a finite-error comparison of bounds with different continuity corrections.
Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Bridge50_P20
Formal statement
noncomputable section
namespace QAlgorithms.Papers.Independent.Bridge50.P20
open QAlgorithms QAlgorithms.Entanglement
-- Reuse pinned accepted definitions; source-specific conventions are audited below.
open QAlgorithms.Papers.Independent.Dependency50.P35 (maxRelative)
open QAlgorithms.Papers.Independent.Information100.P04 (bellProjector)
open QAlgorithms.Papers.Independent.Information100.P53 (amplitudeDamping)
open QAlgorithms.Papers.Independent.Information100.P75 (mixedIdentity)
open QAlgorithms.Papers.Independent.Next46.P29 (zRotation)
open QAlgorithms.Papers.Independent.Support50.P11 (choiState)
open scoped Matrix ComplexConjugate ComplexOrder
attribute [local instance] Classical.propDecidable Classical.decEq
/-- arXiv:1707.05543v2, PDF p.16, (49): the relative advantage lies in [-1,1]
and is positive exactly when the versatile cut cost is strictly smaller.
The zero-cost pair is excluded by the positive-sum hypothesis. -/
theorem advantage_criterion (a b : ℝ) (ha : 0 ≤ a) (hb : 0 ≤ b)
(hsum : 0 < a + b) :
(-1 ≤ relativeAdvantage a b ∧ relativeAdvantage a b ≤ 1) ∧
(0 < relativeAdvantage a b ↔ b < a) := by sorry
end QAlgorithms.Papers.Independent.Bridge50.P20
end
Source
Versatile relative entropy bounds for quantum networks, arXiv:1707.05543v2, PDF p. 16, §IV B, (49): relative advantage parameter
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For all real numbers satisfying , , and , the quantity satisfies
Either or may be zero, but they cannot both be zero.
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.