Corollary A.5
ProvedQAlgorithms.Papers.Independent.Support50.P12.corollary_A_5Let be positive semidefinite operators on a nonzero finite-dimensional complex Hilbert space. Operator powers are zero on kernels; scalar trace powers have their usual real-power convention.
One has
If , the inequality is reversed for every , and
Zero operators are included. The endpoint inequalities do not have a proportionality characterization.
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]
/-- Corollary A.5, PDF pp.45–46: both full exponent ranges and the 'some positive
order other than one' equality characterization, including zero operators. -/
theorem corollary_A_5 [Nonempty I] (X Y : Mat I) (hX : X.PosSemidef) (hY : Y.PosSemidef) :
(∀ a ∈ Icc (0 : ℝ) 1, tr (supportPower X a * supportPower Y (1 - a)) ≤
Real.rpow (tr X) a * Real.rpow (tr Y) (1 - a)) ∧
(Supports X Y → ∀ a ≥ (1 : ℝ), tr (supportPower X a * supportPower Y (1 - a)) ≥
Real.rpow (tr X) a * Real.rpow (tr Y) (1 - a)) ∧
(Supports X Y →
((∃ a : ℝ, 0 < a ∧ a ≠ 1 ∧ tr (supportPower X a * supportPower Y (1 - a)) =
Real.rpow (tr X) a * Real.rpow (tr Y) (1 - a)) ↔
∃ c : ℝ, 0 ≤ c ∧ X = (c : ℂ) • 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 a finite nonempty index type with decidable equality and let be positive semidefinite. Write and for functional calculus with scalar value when and value at , so that is its support projection. Then, for every real ,
If , the reverse inequality holds for every real , and under that same support hypothesis,
No nonzero or trace-normalization assumptions are imposed. The scalar powers on the right are ordinary total real powers, with and for , whereas the matrix powers assign zero to a zero eigenvalue even at exponent .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.