Remark 5.7
OpenQAlgorithms.Papers.Independent.Support50.P12.remark_5_7Let and be concrete unital algebras on nonzero finite-dimensional complex Hilbert spaces, and let be complex linear. The adjoint is for . A substochastic map has a Schwarz-contracting adjoint; stochastic additionally means trace preserving. Let . Powers are zero on kernels and traces are not normalized. Write for the unnormalized Petz Rényi entropy, , , and . Logarithms are natural and . The Petz map is .
The ten-condition result of Theorem 5.1 holds for co-substochastic maps with co-stochastic/co-substochastic recoveries, after replacing its cocycle by
The quantifiers are all complex , one real , or all , in conditions (vi)–(viii), respectively. All other conditions and the asymmetric implication chain remain as stated there; two-copositivity supplies the recovery converse. Recovery may retain each given copositivity level or complete copositivity. The proof's adjoint slips are not adopted.
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P12
noncomputable section
namespace QAlgorithms.Papers.Independent.Support50.P12
open Filter MeasureTheory Set
open QAlgorithms.Papers.Independent.Information100.P60 (Mat copies)
open QAlgorithms.Papers.Independent.Information100.P30 (MatrixMap)
open QAlgorithms.Papers.Independent.Dependency50.P28 (tr LE logNonnegativeExtended pinch)
open QAlgorithms.Papers.Independent.Dependency50.P21 (channelAdjoint spectralFunction petz relativeEntropy)
open QAlgorithms.Papers.Independent.Frontier18.P17 (hs)
open QAlgorithms.Papers.Independent.Next46.P23 (Supports)
open QAlgorithms.Papers.Independent.Support21.P03 (ExtendedConvex)
open QAlgorithms (spectralProjector)
open QAlgorithms.Learning (IsDistribution)
open QAlgorithms.Papers.Independent.Information100.P04 (absOp)
open scoped BigOperators Matrix ComplexConjugate ComplexOrder ENNReal Topology Kronecker
attribute [local instance] Classical.propDecidable Classical.decEq
open QAlgorithms.Papers.Independent.Information100.P71 (supportPower)
variable {I J K : Type} [Fintype I] [DecidableEq I] [Fintype J] [DecidableEq J]
[Fintype K] [DecidableEq K]
/-- Remark 5.7, PDF p.26: co-version of the full chain, with reversed products in
(vi)–(viii) and a 2-copositive converse. The proof's Phi/Phi* slips are not copied. -/
theorem remark_5_7 [Nonempty I] [Nonempty J] (A : Alg I) (B : Alg J) (Phi : MatrixMap I J)
(hp : CoSubstochastic A B Phi) (X Y : Mat I) (hX : X ∈ A) (hY : Y ∈ A)
(hXp : X.PosSemidef) (hYp : Y.PosSemidef) (htr : tr (Phi Y) = tr Y) :
let c := pairConditions true A B Phi X Y
(Supports X Y → PairChain c) ∧ (c 1 → c 2) ∧
(Supports X Y → KCoPositive A Phi 2 → c 9 → c 0) ∧
(Supports X Y → c 0 →
(∀ n, KCoPositive A Phi n → ∃ Psi : MatrixMap J I, CoStochastic B A Psi ∧
KCoPositive B Psi n ∧ Psi (Phi X) = X ∧ Psi (Phi Y) = Y) ∧
((∀ n, KCoPositive A Phi n) → ∃ Psi : MatrixMap J I, CoStochastic B A Psi ∧
(∀ n, KCoPositive B Psi n) ∧ Psi (Phi X) = X ∧ Psi (Phi Y) = Y)) := by sorry
end QAlgorithms.Papers.Independent.Support50.P12
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be nonempty finite index types with decidable equality, , unital complex star subalgebras, and complex linear. Write and . Call a complex-linear co-substochastic from to when its entire range lies in , the entire range of lies in , is positive on positive elements of , and for every ; call it co-stochastic if, additionally, for every . Assume is co-substochastic and are positive semidefinite with . Define by functional calculus with value zero at zero and otherwise, , and by the scalar function zero at zero and elsewhere when is Hermitian, with value zero for a non-Hermitian matrix. Let be the distinct Hermitian spectrum and the empty set otherwise, and the orthogonal eigenspace projection. Define
using extended real values, the total selected-limit operation, and zero for division by zero. A permitted generator is continuous on , has converging in the extended real line at infinity, and is operator convex on positive semidefinite matrices in every size. Let be the set of ratios , with , , for which some nonzero satisfies , and let . A generator is rich here when its selected representing measure on has support cardinality at least ; this is the measure selected, when possible, to satisfy for some real , ,
with and integrable displayed kernels for every , where away from zero and is its selected right limit; the measure choice remains total if no representation exists. Define by
and . Let mean, respectively: existence of a co-stochastic recovery of both ; existence of a co-substochastic recovery of both; divergence equality for every permitted generator; equality for some rich permitted generator; equality for for all in some subset of of cardinality at least ; for every complex ; for some , ; for every real ;
and . Under , the theorem asserts
and it asserts without support inclusion. Define -co-positivity on to mean that, whenever a positive semidefinite block matrix has each transposed block in , applying to each block produces a positive semidefinite matrix. Support inclusion, -co-positivity, and imply . Support inclusion and further imply that every for which is -co-positive admits a co-stochastic recovery that is -co-positive and recovers both ; if is -co-positive for every natural , one co-stochastic recovery has this property for every and recovers both. The quantifiers include , and no nonzero or normalization assumption is imposed on .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.