Incremental learning with sample queries

Joel Ratsaby

Research output: Contribution to journalArticlepeer-review

21 Scopus citations

Abstract

The classical theory of pattern recognition assumes labeled examples appear according to unknown underlying class conditional probability distributions where the pattern classes are picked randomly in a passive manner according to their a priori probabilities. This paper presents experimental results for an incremental nearest-neighbor learning algorithm which actively selects samples from different pattern classes according to a querying rule as opposed to the a priori probabilities. The amount of improvement of this query-based approach over the passive batch approach depends on the complexity of the Bayes rule.

Original languageEnglish
Pages (from-to)883-888
Number of pages6
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume20
Issue number8
DOIs
StatePublished - 1 Dec 1998
Externally publishedYes

Keywords

  • Active learning
  • Incremental learning
  • Model selection
  • Nearestneighbor algorithm
  • Sample querying

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Incremental learning with sample queries'. Together they form a unique fingerprint.

Cite this