Definitions for Additivity for a class of unital qubit channels
DefinitionQAlgorithms_Papers_Independent_Support50_P13The objects the formalized statements of Additivity for a class of unital qubit channels (arXiv:quant-ph/0103156) are stated in. Each was checked against the paper by a blind read-back and a moderator.
ultimateCapacityNats: PDF p.4, Remark 2: operational unassisted classical capacity, converted from bits to nats. The accepted code model allows arbitrary block input states and collective POVMs, and provides no prior sender-receiver entanglement.normalizedHolevoRate: PDF p.4, (8): normalized Holevo quantity of the actual n-use tensor channel. The n=0 extension is zero and is irrelevant to the ordinary limit at infinity.
import Mathlib
import Definitions.Def_QAlgorithms_Papers_Independent_Gap48_P03
import Definitions.Def_QAlgorithms_Papers_Independent_Dependency50_P23
import Definitions.Def_QAlgorithms_Papers_Independent_Next46_P42
/-!
King, arXiv:quant-ph/0103156v2, Additivity for unital qubit channels.
Page numbers are PDF pages. Entropies and classical capacities are in nats.
Imports are pinned accepted libraries. All new definitions are paper-local.
Theorem 2 is an exact specialization of the accepted phaseDampingPNormBound;
its reuse record is in items.json, so it has no duplicate theorem here.
-/
noncomputable section
open scoped Matrix ComplexConjugate ComplexOrder
open Filter
namespace QAlgorithms.Papers.Independent.Support50.P13
open QAlgorithms.Papers.Independent.Gap48.P03
(productChannel maximumOutputPNorm holevoQuantity depolarizingNormFactor
phaseChannel qubitPauli)
open QAlgorithms.Papers.Independent.Next46.P42 (minimumOutputEntropy)
open QAlgorithms.Papers.Independent.Information100.P63 (schatten)
open QAlgorithms.Papers.Independent.Dependency50.P23 (ordinaryClassicalCapacity)
attribute [local instance] Classical.propDecidable Classical.decEq
/-- PDF p.4, Remark 2: operational unassisted classical capacity, converted from
bits to nats. The accepted code model allows arbitrary block input states and
collective POVMs, and provides no prior sender-receiver entanglement. -/
def ultimateCapacityNats {A B : Type} [Fintype A] [Fintype B]
[DecidableEq A] [DecidableEq B] (Phi : Superop A B) : ℝ :=
Real.log 2 * ordinaryClassicalCapacity Phi
/-- PDF p.4, (8): normalized Holevo quantity of the actual n-use tensor channel.
The n=0 extension is zero and is irrelevant to the ordinary limit at infinity. -/
def normalizedHolevoRate {A B : Type} [Fintype A] [Fintype B]
[DecidableEq A] [DecidableEq B] (Phi : Superop A B) (n : ℕ) : ℝ :=
holevoQuantity (piSuperop (fun _ : Fin n => Phi)) / (n : ℝ)
/-- Converting the operational capacity back from nats recovers the bit value. -/
example {A B : Type} [Fintype A] [Fintype B]
[DecidableEq A] [DecidableEq B] (Phi : Superop A B) :
ultimateCapacityNats Phi / Real.log 2 = ordinaryClassicalCapacity Phi := by
have hlog : Real.log (2 : ℝ) ≠ 0 := ne_of_gt (Real.log_pos (by norm_num))
unfold ultimateCapacityNats
field_simp [hlog]
/-- The zero-length extension has the declared harmless value. -/
example {A B : Type} [Fintype A] [Fintype B]
[DecidableEq A] [DecidableEq B] (Phi : Superop A B) :
normalizedHolevoRate Phi 0 = 0 := by
simp [normalizedHolevoRate]
end QAlgorithms.Papers.Independent.Support50.P13
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
QAlgorithms.Papers.Independent.Support50.P13.normalizedHolevoRate
For any finite types , each equipped with decidable equality, any function , and any natural number , this defines
The supremum ranges over every finite ensemble , with arbitrary natural size , real weights summing to , and positive semidefinite matrices of trace indexed by ; zero weights are allowed, while cannot satisfy the weight condition. The function is evaluated by continuous functional calculus, with the total real logarithm satisfying . Explicitly, for ,
where is the corresponding matrix unit. No channel or linearity assumption is made on , and the finite types may be empty. At the defined rate is , because real division by zero gives . The real supremum is also total, giving when its defining set is empty or unbounded above.
QAlgorithms.Papers.Independent.Support50.P13.ultimateCapacityNats
For any finite types , each equipped with decidable equality, and any function , this defines the real number , where is the natural logarithm and is the supremum of all real satisfying the following condition: there is a sequence of positive integers such that and, for every , eventually . Here
where the input matrices are indexed by , the measurement matrices by , a density matrix means a positive semidefinite matrix of trace , and the function is defined on every matrix by
The matrix has entry at and zero elsewhere, and . No channel, linearity, positivity, or nonemptiness assumption is imposed on . The sequences include , with empty products equal to and division by zero defined to give ; all displayed real suprema use the total real supremum, which gives for an empty set or a set unbounded above.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.