Unsupervised Learning for Signal Versus Noise

A. F.M. Smith, Udi E. Makov

Research output: Contribution to journalArticlepeer-review


The Bayes solution to the unsupervised sequential learning problem induced by a mixture model for the two-class signal versus noise decision problem generates a computational and storage explosion. A quasi-Bayes approximate learning procedure is proposed that avoids the computational explosion while retaining the flavor of the Bayes solution. Convergence is established and efficiency is investigated.

Original languageEnglish
Pages (from-to)498-500
Number of pages3
JournalIEEE Transactions on Information Theory
Issue number4
StatePublished - Jul 1981
Externally publishedYes

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences


Dive into the research topics of 'Unsupervised Learning for Signal Versus Noise'. Together they form a unique fingerprint.

Cite this