Additivity of optimized accessible equivocation
OpenQAlgorithms.Papers.Independent.Intakenext100.P63.optimized_accessible_additivityLet and be normalized density matrices on nonzero finite-dimensional spaces and , where and are finite nonempty label sets. These are the output families of two cq-channels. For any family and prior , write , with joint probabilities and arbitrary finite POVMs. Then
The left optimization includes every correlated joint prior, and Eve may measure jointly. Zero prior weights are allowed. This is the additivity statement for complements of cq-channels, expressed directly through their cq output families; all logarithms are binary.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P63
noncomputable section
namespace QAlgorithms.Papers.Independent.Intakenext100.P63
open QAlgorithms Filter
open QAlgorithms.Papers.Independent.Information100.P63 (Mat)
open QAlgorithms.Papers.Independent.Information100.P04 (Measurement)
open QAlgorithms.Papers.Independent.Information100.P27 (shannon labelInformation)
open QAlgorithms.Papers.Independent.Information100.P71 (channelPower dilationMap SupportIncluded)
open QAlgorithms.Papers.Independent.Bridge50.P47 (SymmetricIndex universalSymmetricDilation)
open QAlgorithms.Learning (IsDistribution)
open scoped Matrix ComplexConjugate ComplexOrder
attribute [local instance] Classical.propDecidable Classical.decEq
/-- PDF pp.7–8, Lemma 4: all correlated priors and all joint POVMs are allowed.
Every normalized state family is the output family of a cq-channel. -/
theorem optimized_accessible_additivity {I K E F : Type}
[Fintype I] [Nonempty I] [Fintype K] [Nonempty K]
[Fintype E] [DecidableEq E] [Nonempty E]
[Fintype F] [DecidableEq F] [Nonempty F]
(rho : I → Mat E) (sigma : K → Mat F)
(hrho : ∀ i, IsDensityMatrix (rho i)) (hsigma : ∀ k, IsDensityMatrix (sigma k)) :
optimizedEquivocation (fun x : I × K => Matrix.kronecker (rho x.1) (sigma x.2)) =
optimizedEquivocation rho + optimizedEquivocation sigma := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P63
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be finite nonempty types, with decidable equality on and , and let and be families of complex matrices on and , respectively, each matrix positive semidefinite with trace . For any such family , define
Here the infimum ranges over all finite measurements with a positive number of outcomes, and logarithms and division use their total real-valued conventions, including and division by zero equal to zero. Then
The supremum on the left ranges over all probability distributions on .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.