Lemma 5
OpenQAlgorithms.Papers.Independent.Gap48.P18.extreme_boundary_approximating_measuresFor a nonempty closed convex finite-capacity set , there exist Borel probability laws and concentrated on such that weakly and
This supplies approximating laws on the closed extreme boundary. Without regularity, no optimality of the limiting law is asserted.
import Definitions.Def_QAlgorithms_Papers_Independent_Gap48_P18
noncomputable section
open MeasureTheory Filter
open scoped ENNReal NNReal Topology Matrix.Norms.L2Operator
open QAlgorithms.Papers.Independent.Dependency50.P33
(State PureState traceSize traceClosure IsTraceClassChannel productCoeff)
open QAlgorithms.Papers.Independent.Frontier18.P07
(GeneralizedEnsemble HasFiniteSupport HasAverage averageCoefficients
extendedEntropyNats extendedRelativeEntropyNats
finiteEnsembleCapacity pureStates)
open QAlgorithms.Papers.Independent.Frontier18.P01 (traceFirst traceSecond)
namespace QAlgorithms.Papers.Independent.Gap48.P18
attribute [local instance] Classical.propDecidable Classical.decEq
/-- Lemma 5, arXiv:quant-ph/0510073v2, PDF pp.40.
PDF overlines verified by rendering p.40. -/
theorem extreme_boundary_approximating_measures {I : Type} [Countable I] [Nonempty I]
(A : Set (State I)) (hA : A.Nonempty) (hclosed : IsClosed A)
(hconvex : CoefficientConvexStates A) (hfinite : setChiCapacity A < ⊤) :
∃ (seq : ℕ → GeneralizedEnsemble I) (mu : GeneralizedEnsemble I),
(∀ n, MeasureConcentratedOn (seq n : Measure (State I)) (extremeClosure A)) ∧
MeasureConcentratedOn (mu : Measure (State I)) (extremeClosure A) ∧
Tendsto seq atTop (𝓝 mu) ∧ probabilityAverage mu = optimalAverage A hA ∧
HConverges (fun n => probabilityAverage (seq n)) (optimalAverage A hA) ∧
Tendsto (fun n => measureHolevo (seq n : Measure (State I))) atTop (𝓝 (setChiCapacity A)) := by sorry
end QAlgorithms.Papers.Independent.Gap48.P18
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be a countable, nonempty type, possibly finite. A state on is a complex matrix admitting an expansion , where and , such that every finite quadratic form is real and nonnegative, its real diagonal is summable, and . Write for the infimum of over all such expansions of , with value if there are none; the state topology is generated by the sets , for states and real . For arbitrary coefficient matrices define
where are the corresponding principal submatrices, matrix logarithms use continuous functional calculus, and , viewed in ; the empty finite set is included. For a probability measure on the Borel state space, put and , and let be the value of the fixed classically chosen continuous barycenter map, whose entries are these integrals and whose coordinate functions are integrable. For a state set , let be the supremum of over probability measures whose topological support is finite and contained in . Let be a classically chosen state satisfying for every , if such a state exists, and otherwise the member of chosen from its given nonemptiness hypothesis. Suppose is nonempty and closed, its coefficient matrices form a convex set over , and . Let be the topological closure of the states whose coefficient matrices are extreme points of that coefficient set. Then there exist probability measures , indexed by every , and a probability measure , such that for every , , in the probability-measure topology,
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.