acceptodds
Preprint in the OpenAI Math release

Microscopic jamming in the negative spherical perceptron

OpenAI

Abstract

We prove a sharp feasibility threshold and limiting gap and force laws for the spherical perceptron with margin −1 and quadratic penalty. The limits are taken successively in system size, inverse temperature, and density approaching the threshold from above. They agree for Gaussian coordinates and for the equal mixture of centered Gaussian coordinates with variances and , for every sufficiently small fixed ε. The contact-removed gap cumulative law and the mean-one force cumulative law satisfy as and , with , , and . A finite numerical certificate for these exponent intervals, together with its mathematical error bounds, is included.

open until 1 Jan 2028

est. 50% chance this result is independently verified by the end of 2027.

Not verified 50%Verified 50%

What do you think this paper will get?

All positions stay anonymous.

Discussion (0)

Sign in to comment.