Section 5.1: LAA constants for linear combinations of marginal entropies
OpenQAlgorithms.Papers.Independent.Support50.P18.entropy_combination_LAAConsider finitely many separable quantum systems , selected subsystems , and real coefficients . Set on the real-valued domain where every marginal entropy with nonzero coefficient is finite. Put and . For states in that domain and , their mixture also belongs to the domain and
Thus these sums are valid LAA constants; the source allows smaller constants as well. The statement applies in finite dimension and on the specified finite-entropy domain in infinite dimension.
import Definitions.Def_QAlgorithms_Papers_Independent_Support50_P18
noncomputable section
open Filter
open scoped ENNReal NNReal BigOperators Topology
attribute [local instance] Classical.propDecidable Classical.decEq
namespace QAlgorithms.Papers.Independent.Support50.P18
open QAlgorithms.Papers.Independent.Dependency50.P33 (State PureState traceSize rankOne)
open QAlgorithms.Papers.Independent.Frontier18.P01 (convexOperatorMixture traceSecond)
open QAlgorithms.Papers.Independent.Frontier18.P07
(extendedEntropyNats extendedRelativeEntropyNats)
open QAlgorithms.RealRegisters (Hilbert)
open QAlgorithms.Papers.Independent.Claimed50.P39 (g)
open QAlgorithms.Papers.Independent.Renew46.P08
(coefficientEnergy entropySupremum extendedHalfTraceDistance)
/-- Section 5.1, PDF pp.9--10: negative coefficients contribute to the lower
LAA defect and positive coefficients to the upper defect. Finiteness describes
the real combination's domain in the source's infinite-dimensional discussion.
The mixed marginal entropies are finite as a conclusion. Footnote 5 allows
smaller valid constants; the displayed sums themselves are valid constants. -/
theorem entropy_combination_LAA {n m : ℕ} {I : Fin n → Type}
[∀ i, Countable (I i)] (S : Fin m → Finset (Fin n)) (c : Fin m → ℝ)
(rho sigma : State ((i : Fin n) → I i))
(hrho : ∀ k, c k ≠ 0 → extendedEntropyNats (subsystemMarginal (S k) rho.coefficients) < ⊤)
(hsigma : ∀ k, c k ≠ 0 → extendedEntropyNats (subsystemMarginal (S k) sigma.coefficients) < ⊤)
(p : ℝ) (hp : 0 < p) (hp1 : p < 1) :
(∀ k, c k ≠ 0 → extendedEntropyNats (subsystemMarginal (S k)
(convexOperatorMixture p rho.coefficients sigma.coefficients)) < ⊤) ∧
-(∑ k, max (-c k) 0) * Real.binEntropy p ≤
entropyCombination S c (convexOperatorMixture p rho.coefficients sigma.coefficients) -
p * entropyCombination S c rho.coefficients -
(1 - p) * entropyCombination S c sigma.coefficients ∧
entropyCombination S c (convexOperatorMixture p rho.coefficients sigma.coefficients) -
p * entropyCombination S c rho.coefficients -
(1 - p) * entropyCombination S c sigma.coefficients ≤
(∑ k, max (c k) 0) * Real.binEntropy p := by sorry
end QAlgorithms.Papers.Independent.Support50.P18
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Let be arbitrary natural numbers, let each , , be a countable type, let and for , and let be states on . Here a state is a complex matrix admitting an expansion with and , such that every finite quadratic form is real and nonnegative, its real diagonal entries form a summable family, and their sum is one. For any matrix , write , with the sum over complementary assignments and with a nonsummable complex sum assigned zero. Define
where is a finite coordinate restriction, denotes continuous functional calculus, , and . Suppose and are finite whenever . For every real with , put and . Then is finite whenever , and
No finiteness hypothesis or conclusion is required for terms with . The statement includes and ; in the latter case both bounds and the intervening expression are zero.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.