A Parametrization Scheme for Classifying Models of Learnability

Abstract

We present a systematic framework for classifying, comparing and defining models of computational learnability. Apart from the obvious ‘uniformity’ parameters we present a novel ‘solid learnability’ notion that captures the difference between ‘Guess and Test’ learning algorithms and learnability notions for which consistency with the samples guarantees success.

Cite

Text

Ben-David et al. "A Parametrization Scheme for Classifying Models of Learnability." Annual Conference on Computational Learning Theory, 1989. doi:10.1006/inco.1995.1094

Markdown

[Ben-David et al. "A Parametrization Scheme for Classifying Models of Learnability." Annual Conference on Computational Learning Theory, 1989.](https://mlanthology.org/colt/1989/bendavid1989colt-parametrization/) doi:10.1006/inco.1995.1094

BibTeX

@inproceedings{bendavid1989colt-parametrization,
  title     = {{A Parametrization Scheme for Classifying Models of Learnability}},
  author    = {Ben-David, Shai and Benedek, Gyora M. and Mansour, Yishay},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {1989},
  pages     = {285-302},
  doi       = {10.1006/inco.1995.1094},
  url       = {https://mlanthology.org/colt/1989/bendavid1989colt-parametrization/}
}