Non-Gaussianity from large quantum width
ProvedQAlgorithms.Papers.Independent.Intakenext100.P88.hybrid_large_widthConsider any hybrid layer at Gaussian initialization. Its incoming quantum outputs are , with normalized feature states and real-linearly independent Hermitian observables. The quantum circuit, features and observables may be drawn from an ensemble. Every is measurable and has finite fourth moment under the actual probability law. The classical preactivations are . All real weights and biases are centered Gaussian, have independent coordinates, are jointly independent of the incoming quantum outputs, and satisfy and , with .
Define . In contrast, define , which subtracts only one product of pair moments.
Fix and a width-independent before choosing the width, data space or ensemble. Suppose, for every data quadruple and every ,
The quantification includes all three data pairings. Then
Thus the connected classical four-point function is uniformly . This is a statement at initialization, applicable to each hybrid layer satisfying the hypotheses; it does not imply full Gaussian-process convergence or derivative suppression during training.
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P88
noncomputable section
namespace QAlgorithms.Papers.Independent.Intakenext100.P88
open scoped BigOperators ComplexConjugate Topology NNReal ComplexOrder
open MeasureTheory ProbabilityTheory Filter
/-- PDF p.8 Theorem 5, SM p.38 Eqs.(202)-(203): uniform 1/m suppression.
CW and B precede all widths and ensembles. Equation (203) prints Q on the left; it must
be C, as its preceding Eq.(200) and the contraction show. Only that label is corrected;
no Gaussian-process convergence or derivative/dynamical suppression is asserted. -/
theorem hybrid_large_width (CW CB : ℝ≥0) (B : ℝ) (hB : 0 ≤ B) :
∀ {Ω D R : Type*} [MeasurableSpace Ω] [Fintype R] [DecidableEq R]
(μ : Measure Ω) [IsProbabilityMeasure μ] {h L m : ℕ}
(S : HybridLayer μ D R h L m CW CB),
(∀ (a b c d : D) (j k : Fin m),
|productCovariance μ S.quantum a b c d j k| ≤ B * kd j k) →
∀ (a b c d : D) (r s t u : R),
|connectedFour μ (preactivation S a r) (preactivation S b s)
(preactivation S c t) (preactivation S d u)| ≤
(CW : ℝ) ^ 2 * B / (m : ℝ) *
(kd r s * kd t u + kd r t * kd s u + kd r u * kd s t) := by sorry
end QAlgorithms.Papers.Independent.Intakenext100.P88
end
Read-back
What the Lean code literally says, in plain math · gpt-6.1-sol
Fix nonnegative real numbers and a real number . For every probability space , every type , every finite type with decidable equality, and every natural , consider any supplying and, for each , , unitary matrices , Hermitian matrices , real parameters , unit vectors , and Hermitian observables that are linearly independent over . Set in increasing index order and , and assume all are measurable and in . Suppose also supplies measurable, mutually independent real variables and , with mean-zero Gaussian laws of variances and , whose entire family is independent of the entire family . Write , for integration against , and for when and otherwise. Assume that, for every and ,
thus the expression inside the absolute value is zero when . Then, for every and ,
This is a bound for every such positive finite width . The cases , zero Gaussian variances, and are included; is excluded, and empty or makes the corresponding universal conclusion vacuous.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.