Section 2.1 / (3),(19): change of reference units
OpenQAlgorithms.Papers.Independent.Gap20.P07.referenceUnitsLet be any CPTP channel between nonzero complex Hilbert spaces, and let be integer reference dimensions. Write for Definition 2.1 capacity relative to an ideal -level channel. Then
The scalar product is taken in the extended nonnegative reals. Changing the nontrivial reference dimension changes only the unit of capacity; dimension one is excluded from the logarithmic ratio.
import Definitions.Def_QAlgorithms_Papers_Independent_Gap20_P07
noncomputable section
open Filter MeasureTheory
open scoped Topology ENNReal NNReal Matrix ComplexOrder ComplexConjugate
attribute [local instance] Classical.propDecidable
attribute [local instance] Classical.decEq
attribute [local instance] IsStarNormal.instContinuousFunctionalCalculus
namespace QAlgorithms.Papers.Independent.Gap20.P07
open QAlgorithms
open QAlgorithms.RealRegisters (Hilbert)
open QAlgorithms.Papers.Independent.Dependency50.P33
(State IsTraceClass IsTraceClassChannel traceSize rankOne tensorApply)
open QAlgorithms.Papers.Independent.Frontier30.P14
(Operation isometryCoeff conjugateKernel)
open QAlgorithms.Papers.Independent.Claimed50.P35
(stabilizedTraceClassNorm traceClassTensorMap traceClassChannelPower
transpose transposeBound TraceLinear tracePair MeasurePrepare)
open QAlgorithms.Papers.Independent.Gap48.P29 (fe FiniteCPMap)
/-- PDF pp.3,9, (3),(19): only nontrivial reference systems have a log ratio. -/
theorem referenceUnits {I J : Type} [Nonempty I] [Nonempty J]
(T : Operation I J) (hT : IsTraceClassChannel T)
(n m : ℕ) (hn : 2 ≤ n) (hm : 2 ≤ m) :
relativeCapacity T (id : Operation (Fin n) (Fin n)) =
ENNReal.ofReal (Real.logb 2 m / Real.logb 2 n) *
relativeCapacity T (id : Operation (Fin m) (Fin m)) := by sorry
end QAlgorithms.Papers.Independent.Gap20.P07
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be arbitrary nonempty types and a map from complex coefficient matrices on to those on . Here a trace-class coefficient matrix has a countable expansion with and ; write for the infimum of these sums in . A density coefficient matrix is trace class, has real nonnegative quadratic forms on every finite set of coordinates, and has summable real diagonal entries summing to . Assume preserves trace class, is additive and complex homogeneous there, and preserves density coefficient matrices after extension by the identity on every finite ancillary register, acting on each ancillary matrix block separately. Call maps satisfying these conditions channels. Define . For a target channel , let be the infimum of over channel encoders and decoders of the appropriate types, and define as the supremum of positive real such that every pair of positive-integer sequences , with and , satisfies . Tensor powers apply the map independently to coordinate blocks, and all these infima, suprema, and limits use extended nonnegative values; an empty capacity supremum is . For every natural satisfying and , the assertion is
where is the identity map on complex matrices and the product is taken in . Both logarithms are positive under the stated hypotheses, and infinite capacities are allowed.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.