Learning Programs with an Easy to Calculate Set of Errors

Abstract

Within the study of inductive inference a recurring theme has been to investigate the learning of programs that are not exactly correct. Previous work attempted to quantify the difference between the function to be learned and the one computed by the result of a learning process. In this paper we study a qualitative measure of approximate correctness of the result of attempting to learn a program for a given function. What we require is that the set of errors be somehow easy to describe.

Cite

Text

Gasarch et al. "Learning Programs with an Easy to Calculate Set of Errors." Annual Conference on Computational Learning Theory, 1988. doi:10.3233/FI-1992-163-409

Markdown

[Gasarch et al. "Learning Programs with an Easy to Calculate Set of Errors." Annual Conference on Computational Learning Theory, 1988.](https://mlanthology.org/colt/1988/gasarch1988colt-learning-a/) doi:10.3233/FI-1992-163-409

BibTeX

@inproceedings{gasarch1988colt-learning-a,
  title     = {{Learning Programs with an Easy to Calculate Set of Errors}},
  author    = {Gasarch, William I. and Sitaraman, Ramesh K. and Smith, Carl H. and Velauthapillai, Mahendran},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {1988},
  pages     = {242-250},
  doi       = {10.3233/FI-1992-163-409},
  url       = {https://mlanthology.org/colt/1988/gasarch1988colt-learning-a/}
}