The kappa inequalities for the finite-dimensional functional approach

From papers

For (u,w)(0,1)2(u,w)\in(0,1)^2, define

κ(u,w)=(uw)(J(w)J(u))2uwJ(L1(H2(u)+H2(w)uw)),\kappa(u,w)=\frac{(u-w)(J(w)-J(u))}{2}-|u-w|J\left(L^{-1}\left(\frac{H_2(u)+H_2(w)}{|u-w|}\right)\right),

with κ(u,u)=0\kappa(u,u)=0. The kappa inequalities. The following two assertions hold: for u,w(0,1/2]u,w\in(0,1/2], κ(u,w)κ(1u,w)\kappa(u,w)\leq\kappa(1-u,w); and the function (u,w)κ(H21(u),H21(w))(u,w)\mapsto\kappa(H_2^{-1}(u),H_2^{-1}(w)) is jointly convex. These inequalities would yield the lower bound needed to prove that ϕΨ\phi\in\Psi, and hence would imply the balanced case of the most-informative Boolean function conjecture. They are supported by numerical evidence but are not proved in the paper.

Progress summary

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

Sources & referencesView supporting material

Primary source

Zijie Chen, Amin Gohari and Chandra Nair, “A Differential Equation Approach to the Most-Informative Boolean Function Conjecture”, arXiv:2502.10019 (2025).

Solutions 0

No solutions have been posted yet.