Prescribed ancillary overlap and exact cloning by identity evolution
OpenQAlgorithms.Papers.Independent.Intakenext100.P97.zero_error_attainabilityAll states are positive semidefinite, trace-one matrices, including pure and rank-deficient states. Write for squared Uhlmann fidelity, , and .
Let and let be density operators on . For every prescribed real , there exist unit vectors , with , whose pure density operators satisfy
Here is the joint-output relative error from Eq. (14), starting from . This establishes sharpness of the zero lower bound through a constructed ancillary pair.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P97
noncomputable section
open scoped Matrix ComplexConjugate
namespace QAlgorithms.Papers.Independent.Intakenext100.P97
open QAlgorithms.Papers.Independent.Information100.P67
(squaredFidelity sineDistance receiverChannel)
open QAlgorithms.Papers.Independent.Dependency50.P48 (densityPower)
/-- PDF p.4, proof of Theorem (ii), Eqs. (20)--(21).
For each prescribed ancillary root fidelity in `[0,f]`, normalized pure
ancillas on `H tensor H` are constructed existentially; identity evolution
then attains zero error. No arbitrary fixed ancillary pair is asserted to work. -/
theorem zero_error_attainability {d : ℕ} (hd : 1 < d)
(rho1 rho2 : Matrix (Fin d) (Fin d) ℂ)
(hrho1 : IsDensityMatrix rho1) (hrho2 : IsDensityMatrix rho2)
(hdistinct : rho1 ≠ rho2) (phi : ℝ)
(hphi0 : 0 ≤ phi) (hphif : phi ≤ rootFidelity rho1 rho2) :
∃ Y1 Y2 : EuclideanSpace ℂ (Fin d × Fin d),
IsState Y1 ∧ IsState Y2 ∧
traceRight (pureDensity Y1) = rho1 ∧ traceRight (pureDensity Y2) = rho2 ∧
rootFidelity (pureDensity Y1) (pureDensity Y2) = phi ∧
oneToTwoError rho1 rho2 (pureDensity Y1) (pureDensity Y2) 1 = 0 := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P97
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be a natural number, let be positive semidefinite complex matrices of trace , and let be any real number satisfying
There exist vectors , each of Euclidean norm , such that, writing , one has
In addition, with identity evolution on the three -dimensional registers, set , tracing out the last register, and . Then
Matrix square roots use real continuous functional calculus, and division by zero is interpreted as zero. The allowed interval for includes both endpoints.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.