Lemma 4: measurement variation bounded by trace norm
OpenQAlgorithms.Papers.Independent.Intakethird100.P03.measurement_variationLet be a finite family of mutually orthogonal subspaces of a finite complex Hilbert space, and let be density operators. With ,
The family need not span the Hilbert space. The sum is the unhalved variation expression used in the paper.
This bounds measurement errors even when the observable can have no value.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakethird100_P03
noncomputable section
open scoped Matrix ComplexConjugate Kronecker
namespace QAlgorithms.Papers.Independent.Intakethird100.P03
open QAlgorithms.Papers.Independent.Renew46.P01 (cleanVector)
attribute [local instance] Classical.propDecidable
/-- PDF p.12, Lemma 4: sum of absolute probability differences for a possibly
incomplete orthogonal family; this is not half-normalized total variation. -/
theorem measurement_variation {I : Type*} [Fintype I] [DecidableEq I]
(r : ℕ) (O : OrthogonalProjectorFamily I r) (rho gamma : Matrix I I ℂ)
(hrho : IsDensityMatrix rho) (hgamma : IsDensityMatrix gamma) :
(∑ v, |subspaceProbability rho (O.proj v) -
subspaceProbability gamma (O.proj v)|) ≤ traceNorm (rho - gamma) := by sorry
end QAlgorithms.Papers.Independent.Intakethird100.P03
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every finite type with decidable equality, every natural number , every family of complex matrices satisfying , , and when , and every pair of positive semidefinite complex matrices with ,
The square root on the right is defined by continuous functional calculus. The projector family need not sum to the identity. When , the left side is ; if is empty, the trace-one hypotheses cannot be satisfied.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.