Scaling laws in learning of classification tasks

N. Barkai, H. S. Seung, H. Sompolinsky

Research output: Contribution to journalArticlepeer-review

21 Scopus citations

Abstract

The effect of the structure of the input distribution on the complexity of learning a pattern classification task is investigated. Using statistical mechanics, we study the performance of a winner-take-all machine at learning to classify points generated by a mixture of K Gaussian distributions (''clusters'') in RN with intercluster distance u (relative to the cluster width). In the separation limit u1, the number of examples required for learning scales as NKu-p, where the exponent p is 2 for zero-temperature Gibbs learning and 4 for the Hebb rule.

Original languageEnglish (US)
Pages (from-to)3167-3170
Number of pages4
JournalPhysical review letters
Volume70
Issue number20
DOIs
StatePublished - 1993
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • General Physics and Astronomy

Fingerprint

Dive into the research topics of 'Scaling laws in learning of classification tasks'. Together they form a unique fingerprint.

Cite this