An ideal ancilla preserves maximal norms and minimum entropy
ProvedQAlgorithms.Papers.Independent.Support50.P31.idealChannelNormEntropyLet be a completely positive trace-preserving endomorphism on a nonzero finite-dimensional complex Hilbert space, and let be the identity channel on another such space. Write for maximum output Schatten -norm (operator norm at infinity), and for minimum output entropy in natural units, optimizing over all joint density inputs. Then
The ideal register cannot improve these objectives even when its input is entangled with the channel input. The source's separate minus-infinity branch is false and is explicitly deferred.
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P31
noncomputable section
open Filter Asymptotics
open scoped ComplexConjugate ENNReal
namespace QAlgorithms.Papers.Independent.Support50.P31
attribute [local instance] Classical.propDecidable Classical.decEq
open QAlgorithms.Papers.Independent.Information100.P56 (entropy)
open QAlgorithms.Papers.Independent.Information100.P63 (schatten)
open QAlgorithms.Papers.Independent.Gap48.P03 (holevoQuantity)
open QAlgorithms.Papers.Independent.Claimed50.P38 (tensorSuperop)
open QAlgorithms.Papers.Independent.Dependency50.P44 (tensorChannels)
/-- PDF p.5, Lemma (9), valid norm and entropy branches only. The ideal register
is first, a coordinate permutation of Phi tensor Id. False minus-infinity
branch is separately deferred, rather than silently included. -/
theorem idealChannelNormEntropy {I J : Type} [Fintype I] [DecidableEq I] [Nonempty I]
[Fintype J] [DecidableEq J] [Nonempty J] (Phi : Superop I I) (hPhi : IsChannel Phi) :
(∀ p : ℝ≥0∞, 1 ≤ p → nuP p (idTensorSuperop (R := J) Phi) = nuP p Phi) ∧
nuH (idTensorSuperop (R := J) Phi) = nuH Phi := by sorry
end QAlgorithms.Papers.Independent.Support50.P31
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be finite nonempty types with decidable equality, and let be a map on complex matrices with a finite Kraus representation satisfying . Define on matrices indexed by by , where . For a map , let be the supremum over positive semidefinite trace-one inputs of for finite , and of for ; let be the infimum over these inputs of , with natural logarithms and . Then
Both one-dimensional and one-dimensional are included.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.