Theorem 2 / abstract headline: Gaussian REE is attained at a separability border
OpenQAlgorithms.Papers.Independent.Intakenext100.P46.gree_borderLet be an entangled Gaussian density operator on finitely many complete bosonic modes, with at least one mode at each party of a fixed bipartition. All separability statements mean trace-norm closed convex-hull separability. Write for extended Umegaki relative entropy with natural logarithms and for its infimum over all separable Gaussian densities, with arbitrary means and no energy restriction. Then there exists such a density with
and its covariance lies on the separability border relative to physical Gaussian covariances: every covariance neighborhood at fixed mean contains an entangled Gaussian density. Pure normal modes are included. This gives both attainment and the Gaussian border restriction.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P46
noncomputable section
namespace QAlgorithms.Papers.Independent.Intakenext100.P46
open scoped BigOperators ENNReal ComplexOrder Matrix
attribute [local instance] Classical.propDecidable Classical.decEq
open QAlgorithms.RealRegisters (Hilbert)
open QAlgorithms.Papers.Independent.Dependency50.P33 (State IsSeparable)
open QAlgorithms.Papers.Independent.Dependency50.P13
(finiteRelativeEntropy compression entropyWeight)
open QAlgorithms.Papers.Independent.Claimed50.P47
(Occupation bipartiteCharacteristic bipartiteWeyl gaussianStateCharacteristic)
-- These accepted definitions have the same completed Hilbert-space meaning.
open QAlgorithms.Papers.Independent.Next46.P13 (coordinateBasis)
open QAlgorithms.Papers.Independent.Frontier18.P18 (operatorCoeff boundedOperator)
open QAlgorithms.Papers.Independent.Claimed50.P41 (IsDiagonal)
/-- Chen, Theorem 2, PDF pp.4–6 (abstract p.1): Gaussian REE is attained
at a covariance border on the same nontrivial cut. Singular densities remain. -/
theorem gree_border {A B : Type} [Fintype A] [Fintype B]
[Nonempty A] [Nonempty B] (rho : State (Occupation A × Occupation B))
(hgaussian : IsGaussianDensity rho) (hentangled : ¬ IsSeparable rho.coefficients) :
∃ sigma : State (Occupation A × Occupation B), GaussianBorder sigma ∧
operatorRelativeEntropy rho sigma = gree rho := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P46
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For any finite nonempty types , let , and let be a state on . Here a state is a complex matrix admitting a representation , with and , such that every finite quadratic form of is real and nonnegative, its real diagonal is summable, and . Put on the quadrature labels . For a real quadrature vector , the Weyl matrix is the product, over all modes, of matrices with entries
with the - and -mode products combined by multiplying their entries. A state is Gaussian here if there exist a real vector and a real symmetric matrix with positive semidefinite such that, for every , each inner sum and the outer sum defining are summable and
A matrix is separable here if it can be approximated arbitrarily closely by finite convex combinations of products of states on and , where the size of an error matrix is the infimum, over the nuclear representations just described, of , with empty infimum . Define by realizing each coefficient matrix as a bounded operator on , choosing such an operator if one exists and otherwise using zero, and taking the supremum over every and every orthonormal -tuple of
where , , matrix functions use continuous functional calculus, , and . This supremum includes the empty tuple . Write for the infimum of over separable Gaussian states , allowing an empty infimum to be . If is Gaussian and is not separable, then there exists a state with such that is separable and has Gaussian data for which, for every , there exist a nonseparable state and a real matrix having the same Gaussian conditions and characteristic formula with data , and satisfying for every pair of quadrature indices.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.