Arbitrarily many uses can fail to detect positive quantum capacity
OpenQAlgorithms.Papers.Independent.Intakenext100.P70.arbitrarilyManyUsesLet be an integer. For a finite-dimensional quantum channel , write
where the maximization ranges over all positive semidefinite, trace-one joint input matrices, including entangled inputs. Entropies and rates are in bits, and is the unassisted quantum capacity.
There exist positive finite-dimensional input and output registers, a completely positive trace-preserving channel between them, and an integer such that
The register dimensions may depend on . This shows that no fixed finite number of channel uses suffices to detect positive quantum capacity for every channel.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P70
namespace QAlgorithms.Papers.Independent.Intakenext100.P70
noncomputable section
open scoped Matrix ComplexConjugate ComplexOrder
attribute [local instance] Classical.propDecidable Classical.decEq
open QAlgorithms.Papers.Independent.Information100.P22
(Op coherentInformation singleUseInformation quantumCapacity mixed)
/-- Abstract/introduction, PDF pp.1--2; proved on p.11: no finite block length detects
positive capacity for every channel. Dimensions may grow with n; all inputs are optimized. -/
theorem arbitrarilyManyUses (n : ℕ) (hn : 0 < n) :
∃ a b : ℕ, 0 < a ∧ 0 < b ∧ ∃ Phi : Superop (Fin a) (Fin b),
IsChannel Phi ∧ cubittBlockInformation Phi n = 0 ∧
∃ t : ℕ, n < t ∧ 0 < cubittBlockInformation Phi t ∧ 0 < quantumCapacity Phi := by sorry
end
end QAlgorithms.Papers.Independent.Intakenext100.P70
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
For every natural number , there exist positive natural numbers , a map from complex matrices to complex matrices, and a natural number , such that has a finite Kraus representation
and
Here
For a density matrix indexed by a finite set , the purification used here has coordinates ; the extension applies to each matrix block with fixed reference indices. The tensor map is defined entrywise by
Finally, , with applied by continuous functional calculus. Thus the claimed zero concerns precisely the normalized supremum for tensor factors; the declaration makes no separate assertion for each smaller positive number of factors.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.