ML Anthology

The Consistent Concept Axiom

Zhaogang Qian, Keki B. Irani
ICML 1991 pp. 437-441
doi:10.1016/B978-1-55860-200-7.50090-8 /icml/1991/qian1991icml-consistent/

Abstract

In this paper, we introduce a consistent concept axiom, which represents version space as a sentence in a formal language. By representing the version space as an axiom, inductive conclusions become logical consequences of a set of axioms including background knowledge, observations, and the consistent concept axiom. This unifies deduction and induction in the same logical framework, so that learning can be more uniformly integrated with the rest of an AI system which is mainly of deductive nature.

ICML Semantic Scholar

Cite

Text

Qian and Irani. "The Consistent Concept Axiom." International Conference on Machine Learning, 1991. doi:10.1016/B978-1-55860-200-7.50090-8

Markdown

[Qian and Irani. "The Consistent Concept Axiom." International Conference on Machine Learning, 1991.](https://mlanthology.org/icml/1991/qian1991icml-consistent/) doi:10.1016/B978-1-55860-200-7.50090-8

BibTeX

@inproceedings{qian1991icml-consistent,
  title     = {{The Consistent Concept Axiom}},
  author    = {Qian, Zhaogang and Irani, Keki B.},
  booktitle = {International Conference on Machine Learning},
  year      = {1991},
  pages     = {437-441},
  doi       = {10.1016/B978-1-55860-200-7.50090-8},
  url       = {https://mlanthology.org/icml/1991/qian1991icml-consistent/}
}