English

A note on the capacity of the binary perceptron

Probability 2024-01-30 v1 Discrete Mathematics Machine Learning Machine Learning

Abstract

Determining the capacity αc\alpha_c of the Binary Perceptron is a long-standing problem. Krauth and Mezard (1989) conjectured an explicit value of αc\alpha_c, approximately equal to .833, and a rigorous lower bound matching this prediction was recently established by Ding and Sun (2019). Regarding the upper bound, Kim and Roche (1998) and Talagrand (1999) independently showed that αc\alpha_c < .996, while Krauth and Mezard outlined an argument which can be used to show that αc\alpha_c < .847. The purpose of this expository note is to record a complete proof of the bound αc\alpha_c < .847. The proof is a conditional first moment method combined with known results on the spherical perceptron

Cite

@article{arxiv.2401.15092,
  title  = {A note on the capacity of the binary perceptron},
  author = {Dylan J. Altschuler and Konstantin Tikhomirov},
  journal= {arXiv preprint arXiv:2401.15092},
  year   = {2024}
}
R2 v1 2026-06-28T14:28:31.510Z