Gaussian inequalities with different covariance matrices - the Heterovariant Gaussian Brunn-Minkowski inequality

Abstract: We establish the following novel heterovariant Gaussian Brunn-Minkowski inequality (involving general covariance matrices $Σ<em>1,Σ_2,Γ$) for origin-symmetric convex sets $K,L \subset \mathbb{R}<sup>n$ and $p,λ\in (0,1)$: [ γ{Σ1}(K)^{1-p} γ{Σ2}(L)^p \le \left( \left( \frac{1-p}{1-λ} \right)^{1-p} \left( \frac{p}λ \right)^p \right)^{n} γΓ((1-λ) K + λL) , ] whenever $Γ\leq Σ_1 \sharp_p Σ_2$, the $p$-geometric mean of $Σ_1,Σ_2$. This immediately implies both the dimensional Gaussian Brunn-Minkowski inequality of Eskenazis-Moschidis (by using $Σ_1 = Σ_2 = Γ$ and optimizing over $p$) and the strong Gaussian (B)-theorem of Cordero-Erausquin-Fradelizi-Maurey (by using $K=L$, $p=λ$ and diagonal covariances), fusing them into a single unified statement. In fact, the inequality applies to convex sets whose appropriate Gaussian barycenter is at the origin. Furthermore, it reveals a new bilinear version of the strong Gaussian (B)-theorem (involving two convex sets); in particular, for all $a,b &gt; 0$ and $λ\in [0,1]$: [ γ(a K)^{1-λ} γ(b L)^λ \leq γ\left ( a^{1-λ} b^λ\left ((1-λ) K + λL \right) \right ) . ] A more general heterovariant version of the (bilinear) (B)-theorem replaces dilation by general linear transformations. A different version yields a heterovariant form of the recently established Gaussian conjugate Rogers-Shephard inequality. By averaging these Gaussian inequalities with respect to the covariance matrices, we obtain dimensional Brunn-Minkowski and conjugate Rogers-Shephard inequalities for Gaussian mixtures. The proofs are based on the Gaussian Forward-Reverse Brascamp-Lieb inequality established in our previous work.