Square-summable coefficients for adding one quadratic denominator
OpenRybinAI2026.P01.crossIntegral_add_l2_coefficientsLet and be real symmetric positive-definite matrices. There should exist nonnegative constants and , depending only on and , such that
Adding to the first quadratic denominator contracts every mixed spherical integral by relative to denominator and by relative to denominator . The same two constants work when the addition is made in the second sphere variable. Explicitly, for every numerator and every positive-definite other denominator ,
The analogous two inequalities hold for . Here is the mission's crossIntegral with the original unnormalized surface measure.
This one-sphere coefficient statement is a sufficient analytic core for CUHK-Shenzhen AI Math Problem 1. Applying it to and makes the two coefficients multiply across the sphere variables; Cauchy--Schwarz then makes their products sum to at most one.
import Definitions.Def_rybin2026_p01_cross_integral open Matrix RybinAI2026.P01
theorem RybinAI2026.P01.crossIntegral_add_l2_coefficients
{n : ℕ} (A B : Matrix (Fin n) (Fin n) ℝ)
(hA : A.PosDef) (hB : B.PosDef) :
∃ α β : ℝ,
0 ≤ α ∧ 0 ≤ β ∧ α ^ 2 + β ^ 2 ≤ 1 ∧
∀ (X Y P : Matrix (Fin n) (Fin n) ℝ), P.PosDef →
(crossIntegral X Y (A + B) P ≤ α * crossIntegral X Y A P) ∧
(crossIntegral X Y (A + B) P ≤ β * crossIntegral X Y B P) ∧
(crossIntegral X Y P (A + B) ≤ α * crossIntegral X Y P A) ∧
(crossIntegral X Y P (A + B) ≤ β * crossIntegral X Y P B) := by
sorry