Fourth Frobenius moment of a scaled Gaussian matrix:
ProvedGaussianMatrix.frobenius_fourth_momentFix matrices and and let be a standard Gaussian matrix. Then
Since is a quadratic form in the Gaussian vector , this is the classical formula for a symmetric (Isserlis' theorem), with . The special case , is the chi-square moment used in the analysis of early-truncated sketching, and the identity controls the variance of since .
Formalization Note. The expectation is a Lebesgue integral against of a polynomial in the entries of .
import Definitions.Def_GaussianMatrix_basic open MeasureTheory ProbabilityTheory open scoped Matrix
namespace GaussianMatrix
theorem frobenius_fourth_moment {a p m n : ℕ} (S : Matrix (Fin a) (Fin p) ℝ)
(T : Matrix (Fin m) (Fin n) ℝ) :
∫ G, frobSq (S * Matrix.of G * T) ^ 2 ∂(gaussianMatrix p m)
= (frobSq S * frobSq T) ^ 2 + 2 * frobSq (Sᵀ * S) * frobSq (T * Tᵀ) := by sorry
end GaussianMatrixRead-back
What the Lean code literally says, in plain math · claude-fable-5-1
Let be natural numbers (implicit; any may be ), and let and be arbitrary real matrices (no hypotheses on them). Let be the law of a array with independent entries, and write (empty sum ). The statement asserts
where , and are matrix products, so that and . The left-hand side is Lean's Bochner integral of the real function (a degree-4 polynomial in the entries of ); by Lean's convention it is if the integrand is not integrable, and the statement does not separately assert integrability. Edge cases: if or the integrand is identically and the right-hand side involves or together with or ; if or the array space is a single point and .
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.