The multilinear good-fibration Brascamp–Lieb inequality

Let VV be a manifold and let Γ1,,Γn\Gamma_1,\dots,\Gamma_n be a transverse family of good fibrations of VV, with corresponding maps πj:VUj\pi_j:V\to U_j and forms ωj\omega_j. For each xVx\in V, let Φx\Phi_x be the map associated with the corresponding flows, and assume that every Φx\Phi_x is injective. Then, for all nonnegative fjL1(Uj)f_j\in L^1(U_j), the multilinear good-fibration inequality.

Vj=1nfj(πjx)1/(n1)ω1(x)ωn(x)1/(n1)  dxj=1n(Ujfj)1/(n1).\int_V \prod_{j=1}^n f_j(\pi_j x)^{1/(n-1)}\,\omega_1(x)\wedge\dots\wedge\omega_n(x)^{1/(n-1)}\;{\rm d}x\leq\prod_{j=1}^n\left(\int_{U_j}f_j\right)^{1/(n-1)}.

This is a Brascamp–Lieb-type estimate for transverse families of good fibrations under the global injectivity hypothesis on the maps Φx\Phi_x. The supplied text does not state whether the assertion is proved or remains open.

Sources & referencesView supporting material

Primary source

Anthony Carbery, Timo S. Hänninen and Stefán Ingi Valdimarsson, “Multilinear Duality and Factorisation for Brascamp-Lieb-type Inequalities with applications”, arXiv:1809.02449 (2020).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.