Root-fidelity lower bound for arbitrary positive operators
ProvedQAlgorithms.Papers.Independent.Advance50.P05.positiveOperatorFidelityLowerBoundLet be arbitrary positive semidefinite operators on a finite-dimensional Hilbert space. Neither a trace-one condition nor a trace upper bound is assumed. Define root fidelity and the full trace norm by and . Then
This includes zero operators and operators of unequal or arbitrarily large trace. It supplies the unnormalized fidelity estimate used in decoupling arguments.
import Definitions.Def_QAlgorithms_Papers_Independent_Advance50_P05
noncomputable section
namespace QAlgorithms.Papers.Independent.Advance50.P05
open QAlgorithms Filter
open scoped Matrix ComplexConjugate ComplexOrder
open QAlgorithms.Papers.Independent.Dependency50.P19 (first both epr IsIso)
open QAlgorithms.Papers.Independent.Information100.P07 (support)
open QAlgorithms.Papers.Independent.Information100.P71 (supportPower SupportIncluded)
open QAlgorithms.Papers.Independent.Dependency50.P35 (elog marginalAB marginalAC)
open QAlgorithms.Papers.Independent.Support50.P21 (dMin weakPositive PositiveTensorBasis)
open QAlgorithms.Papers.Independent.Information100.P21 (FiniteLOCC)
open QAlgorithms.Papers.Independent.Information100.P67 (IsSubnormalized)
attribute [local instance] Classical.propDecidable Classical.decEq
/-- Buscemi--Datta, arXiv:1006.1896v2, PDF pp.16--17, Lemma 11, (62).
Arbitrary positive operators with arbitrary traces; no normalization or trace cap.
The trace norm is the full Schatten one-norm and F is root fidelity. -/
theorem positiveOperatorFidelityLowerBound
{I : Type} [Fintype I] [DecidableEq I]
(P Q : Matrix I I ℂ) (hP : P.PosSemidef) (hQ : Q.PosSemidef) :
rootFidelity P Q ≥ (P.trace.re + Q.trace.re) / 2 - traceNorm (P - Q) / 2 := by sorry
end QAlgorithms.Papers.Independent.Advance50.P05
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every finite type with decidable equality, possibly empty, and every pair of positive semidefinite complex matrices on , the functional-calculus quantity
satisfies
The last trace is the defined trace norm of . Neither matrix is required to have trace or to be nonzero; for an empty index type all displayed traces are zero.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.