학술논문

Capacity threshold for the Ising perceptron
Document Type
Working Paper
Author
Source
Subject
Mathematics - Probability
Condensed Matter - Disordered Systems and Neural Networks
Mathematical Physics
Language
Abstract
We show that the capacity of the Ising perceptron is with high probability upper bounded by the constant $\alpha_\star \approx 0.833$ conjectured by Krauth and M\'ezard, under the condition that an explicit two-variable function $\mathscr{S}_\star(\lambda_1,\lambda_2)$ is maximized at $(1,0)$. The earlier work of Ding and Sun proves the matching lower bound subject to a similar numerical condition, and together these results give a conditional proof of the conjecture of Krauth and M\'ezard.
Comment: 76 pages, 2 figures