Fidelity squared is an attained primal SDP value
OpenQAlgorithms.Papers.Independent.Intakenext100.P98.fidelity_primal_maximumLet and , where , and let be arbitrary vectors. Define and , and let denote root fidelity. Then
The maximum is attained. Neither vector must have norm one, and the marginal constraint is an inequality. This is the primal fidelity SDP obtained from Uhlmann’s theorem.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P98
noncomputable section
open scoped Matrix ComplexConjugate ComplexOrder
namespace QAlgorithms.Papers.Independent.Intakenext100.P98
/-- PDF p.14, Corollary 7: squared fidelity is a MAXIMUM with a marginal
INEQUALITY; P,Q are constructed from the given, unnormalized vectors. -/
theorem fidelity_primal_maximum {m r : ℕ} (hm : 0 < m) (hr : 0 < r)
(u v : EuclideanSpace ℂ (Fin m × Fin r)) :
IsGreatest (fidelityPrimalValues u v)
((rootFidelity (traceLeft (pureDensity u)) (traceLeft (pureDensity v))) ^ 2) := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P98
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For positive natural numbers and arbitrary vectors , let , , and let sum over the first coordinate: . Set , , and
Then is the greatest element of the set of numbers for matrices satisfying and . Thus such a matrix attains value , and every feasible has value at most . Here is conjugate transpose, means positive semidefinite, and the square roots are positive semidefinite matrix square roots. No normalization or nonzero assumption is imposed on .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.