Equation (9): two-dimensional LOG feasibility criterion
OpenQAlgorithms.Papers.Independent.Gap48.P47.criterionNineLet be normalized two-mode Gaussian oscillator states with finite second moments and arbitrary means. Use , vacuum covariance , and associated normal vectors with . Define
and , . Then
LOG consists of independent finite-ancilla local Gaussian dilations and partial traces on the actual density operators. No restriction is imposed. The positive lower product bound excludes zero divisors.
This is the alternative feasibility criterion before reducing the search to simultaneous zero sets.
import Definitions.Def_QAlgorithms_Papers_Independent_Gap48_P47
noncomputable section
open scoped BigOperators ComplexOrder
namespace QAlgorithms.Papers.Independent.Gap48.P47
open QAlgorithms.Papers.Independent.Dependency50.P33
(State IsTraceClass IsTraceClassChannel)
open QAlgorithms.Papers.Independent.Claimed50.P19
(Cov FockIndex DensityKernel GaussianWithCovariance CharacteristicSummable
multimodeWeylTrace covarianceCharacteristicValue modeCovarianceBlock
standardFormCovariance)
open QAlgorithms.Papers.Independent.Renew46.P39 (AffineMultimodeGaussianUnitary)
open QAlgorithms.Papers.Independent.Frontier30.P08 (interleavedSymplecticForm)
open QAlgorithms.Papers.Independent.Dependency50.P10 (complexify)
open QAlgorithms.Papers.Independent.Frontier30.P14 (boundedUnitaryConjugation)
attribute [local instance] Classical.propDecidable Classical.decEq
/-- PDF p.3, (9): the alternative two-dimensional inequality criterion.
No ordering constraint x>=|y| is inserted. -/
theorem criterionNine (rho target : TwoModeGaussianDensity)
(hsource : 0 < (associatedVector rho.covariance).four)
(htarget : 0 < (associatedVector target.covariance).four) :
LOGReachable rho target ↔
∃ x y : ℝ, 0 < x ∧
w2 rho.covariance target.covariance ≤ |x * y| ∧
|x * y| ≤ w1 rho.covariance target.covariance ∧
0 ≤ f1 rho.covariance target.covariance x y ∧
0 ≤ f2 rho.covariance target.covariance x y := by sorry
end QAlgorithms.Papers.Independent.Gap48.P47
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be any two two-mode Gaussian density objects with respective covariances . For any covariance , write for its mode block and define
Put , , and assume and . Define
Then independent local Gaussian dilations take exactly to if and only if there are real such that
The variable ranges over all real numbers; all divisions use the convention that division by is , and real square roots of negative arguments are . The full meaning of the Gaussian objects and dilation condition is the following. Let , , and with J=\begin{psmallmatrix}0&1\\-1&0\end{psmallmatrix}, in mode/quadrature order. A nuclear kernel admits with summable and . A density kernel additionally has real nonnegative quadratic forms on every finite set of indices and summable real diagonal of sum . A Gaussian density object pairs a density kernel with a real positive definite covariance satisfying , requires summability of , and has a real mean such that for every real ,
where each inner series and the outer series are summable and
The dilation condition means that there exist one-mode kernel maps satisfying
Each of these maps must preserve nuclear kernels, be complex linear on them, and preserve density kernels under slice amplification to for every . Each must also admit its own environment size , Gaussian environment density kernel , and affine Gaussian unitary on , meaning a unitary with real satisfying and coefficientwise , for which
for every nuclear . Here , and all coefficientwise conjugations use this nested-sum convention. Zero environment modes are included, with singleton ; the amplification test has no density inputs.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.