Distribution-Independent Reliable Learning

  title={Distribution-Independent Reliable Learning},
  author={Varun Kanade and Justin Thaler},
We study several questions in the reliable agnosticlearning framework of Kalai et al. (2009), which captures learning tasks in which one type of error is costlie r than other types. A positive reliable classifier is one that makes no false positive errors. The goal in the positive reliableagnostic framework is to output a hypothesis with the following properties: (i) its f alse positive error rate is at most ǫ, (ii) its false negative error rate is at most ǫ more than that of the best positive… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 2 times over the past 90 days. VIEW TWEETS


Publications referenced by this paper.
Showing 1-10 of 28 references

Sherstov . Approximating the andor tree

  • A. Alexander
  • Theory of Computing
  • 2013

Sherstov . Optimal bounds for signrepresenting the intersection of two halfspaces by polynomials

  • A. Alexander
  • Combinatorica
  • 2013

ACM . Ran ElYaniv and Yair Wiener . On the foundations of noisefree selective classification

  • Charles Elkan
  • Journal of Machine Learning Research
  • 2010

On the problem of the most efficient tests for statistical hypotheses

  • J. Neyman, E. S. Pearson
  • Philos . Trans . R . So . Lond . Ser . A Contain…
  • 2010

Sherstov . A separation of np and conp in multiparty communication complexity

  • David Haussler
  • Theory of Computing
  • 2010

Similar Papers

Loading similar papers…