§6.1.4 (6.25)–(6.27): conditioning on a classical register
OpenQAlgorithms.Papers.Independent.Intakenext100.P07.classical_block_entropyAll 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 .
Let p_y>=0 with sum_y p_y<=1 and let each conditional rhohat_AB(y) be normalized. Assemble the actual classical block state sum_y p_y rhohat_AB(y) tensor |y><y|. The displayed min- and max-entropy identities hold, including zero total mass. For any normalized classical joint law, the optimal guessing probability is sum_y max_x p(x,y)=2^(-Hmin(X|Y)).
This result records the source’s operational or structural relation, with the stated endpoint and attainment clauses retained.
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.classicalconditioning
§6.1.4 (6.25)–(6.27): conditioning on a classical register, arXiv:1504.00233v5, PDF p.96. -/
theorem classical_block_entropy [Nonempty A] [Nonempty B] [Nonempty X] [Nonempty Y]
(p : Y → ℝ) (hp : ∀ y, 0 ≤ p y) (ht : ∑ y, p y ≤ 1)
(r : Y → paperMat (A × B)) (hr : ∀ y, IsDensityMatrix (r y)) :
paperHmin (conditionalBlocks (fun y => (p y:ℂ) • r y)) =
-paperLogE (∑ y, p y*(2:ℝ)^(-(paperHmin (r y)).toReal)) ∧
hmax (conditionalBlocks (fun y => (p y:ℂ) • r y)) =
paperLogE (∑ y, p y*(2:ℝ)^(hmax (r y)).toReal) ∧
(∀ q : X × Y → ℝ, paperProbability q →
expE (-paperHmin (Matrix.diagonal (fun xy => (q xy:ℂ)))) =
ENNReal.ofReal (∑ y, sSup {v : ℝ | ∃ x, v = q (x,y)})) := 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 be nonempty finite types with decidable equality, let satisfy and , and let each be a positive semidefinite matrix of complex trace on . For a matrix on , define
where means , functional calculus supplies matrix square roots, , and is the extended-real embedding of for . Form on by
Then
where denotes conversion of an extended real to a real, taking both infinities to . Moreover, for every with and ,
Here the diagonal matrix is viewed as bipartite on , truncates a real number below at and embeds it in , and takes values at and finite , respectively. Zero weights and the all-zero function are included; no normalization is imposed.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.