§6.3.1 (6.58): exact smoothing of flat entropy
OpenQAlgorithms.Papers.Independent.Intakenext100.P07.flat_entropy_smoothingAll quantum systems below are finite-dimensional complex Hilbert spaces of positive dimension, and logarithms are binary. A normalized state is positive semidefinite with trace one; a subnormalized state has trace at most one. Marginals are partial traces. Here is squared algebraic fidelity, is squared generalized fidelity, is purified distance, and . The symbols and denote conditional min- and max-entropy, with values and at zero. Smoothing optimizes over all subnormalized states in the purified-distance ball; every entropy radius satisfies .
For a normalized state rho with equal min- and max-entropy and 0 <= epsilon < 1, both smoothing optima are attained by t=(1-epsilon^2)rho. In bits,
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P07
noncomputable section
open Filter
open scoped BigOperators Matrix ComplexConjugate ComplexOrder Matrix.Norms.L2Operator Topology
namespace QAlgorithms.Papers.Independent.Intakenext100.P07
attribute [local instance] Classical.propDecidable Classical.decEq
variable {I J A B C D X Y Z : Type*}
variable [Fintype I] [DecidableEq I] [Fintype J] [DecidableEq J]
variable [Fintype A] [DecidableEq A] [Fintype B] [DecidableEq B]
variable [Fintype C] [DecidableEq C] [Fintype D] [DecidableEq D]
variable [Fintype X] [DecidableEq X] [Fintype Y] [DecidableEq Y]
variable [Fintype Z] [DecidableEq Z]
/-- NODE paper:1504.00233#6.flatentropy
§6.3.1 (6.58) and its following max-entropy counterpart, PDF p.102.
Both optima are attained by the explicitly constructed (1-e^2)r. -/
theorem flat_entropy_smoothing [Nonempty A] [Nonempty B] (r : paperMat (A × B)) (hr : IsDensityMatrix r)
(hf : paperHmin r = hmax r) (e : ℝ) (he : 0 ≤ e) (he1 : e < 1) :
let t := ((1-e^2:ℝ):ℂ) • r
t ∈ paperBall r e ∧ paperSmoothMin r e = paperHmin t ∧ paperSmoothMin r e = paperHmin r-(Real.logb 2 (1-e^2):EReal) ∧
paperSmoothMax r e = hmax t ∧ paperSmoothMax r e = hmax r+(Real.logb 2 (1-e^2):EReal) := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P07
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let and be finite, nonempty types with decidable equality, and let be a complex matrix indexed by , positive semidefinite with . To specify the quantities in the statement, call a matrix subnormalized when and , and write
where matrix square roots are defined by continuous functional calculus. Define the following extended-real quantities:
where each is indexed by , and matrix inequalities mean that the difference is positive semidefinite. For a real number , define
and define and . The suprema and infimum defining these entropies are taken in the extended reals; the optimization matrices may be zero or have trace strictly below one. If , then for every real satisfying , the matrix satisfies all five assertions
The real logarithms in these equalities are embedded into the extended reals. The permitted case gives ; throughout the stated range, .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.