Sec. II A, unnumbered: convexity of compatible lists and extension fibers
ProvedQAlgorithms.Papers.Independent.Extend50.P36.joining_lists_and_fibers_convexFix finitely many finite-dimensional complex tensor factors, each of positive dimension, and any finite list of neighborhoods. A marginal list is attainable if one density operator reduces to all its prescribed neighborhood states. Then
Convex combinations of lists are componentwise; partial traces are over the exact tensor complements. Empty lists and empty neighborhoods are permitted and cause no missing normalization constraint.
import Definitions.Def_QAlgorithms_Papers_Independent_Extend50_P36
noncomputable section
open scoped Matrix ComplexConjugate Matrix.Norms.Elementwise
open MeasureTheory
namespace QAlgorithms.Papers.Independent.Extend50.P36
attribute [local instance] Classical.propDecidable Classical.decEq
-- Exact accepted matrix alias; tensor reductions below compose frozen superoperators.
open QAlgorithms.Papers.Independent.Information100.P21 (Op)
/-- Sec.II A, PDF p.3: convexity of attainable marginal lists and of every extension fiber. -/
theorem joining_lists_and_fibers_convex (N q : ℕ) (dim : Fin N → ℕ)
(hdim : ∀ i, 0 < dim i) (S : Fin q → Finset (Fin N)) :
Convex ℝ {rho : (j : Fin q) → Op (NeighborhoodConfiguration dim (S j)) | ∃ X, JoiningList dim S rho X} ∧
(∀ rho : (j : Fin q) → Op (NeighborhoodConfiguration dim (S j)),
Convex ℝ {X | JoiningList dim S rho X}) := by sorry
end QAlgorithms.Papers.Independent.Extend50.P36
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For all natural numbers , every family of positive natural dimensions , and every indexed family of subsets , , let . For a matrix indexed by full configurations, define its marginal on by
where takes its coordinates from on and from outside . The set of families of complex matrices , with indexed by , for which there exists a positive semidefinite matrix of trace satisfying for every , is convex over . For every family , the set of all such is also convex over : both sets are closed under for . The subsets may overlap, repeat, be empty, or be the full index set; and are allowed, and a fiber with no such is included as an empty convex set.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.