Lower Bounds for the Complexity of Learning Half-Spaces with Membership Queries

Abstract

Exact learning of half-spaces over finite subsets of ℝ^n from membership queries is considered. We describe the minimum set of labelled examples separating the target concept from all the other ones of the concept class under consideration. For a domain consisting of all integer points of some polytope we give non-trivial lower bounds on the complexity of exact identification of half-spaces. These bounds are near to known upper bounds.

Cite

Text

Shevchenko and Zolotykh. "Lower Bounds for the Complexity of Learning Half-Spaces with Membership Queries." International Conference on Algorithmic Learning Theory, 1998. doi:10.1007/3-540-49730-7_5

Markdown

[Shevchenko and Zolotykh. "Lower Bounds for the Complexity of Learning Half-Spaces with Membership Queries." International Conference on Algorithmic Learning Theory, 1998.](https://mlanthology.org/alt/1998/shevchenko1998alt-lower/) doi:10.1007/3-540-49730-7_5

BibTeX

@inproceedings{shevchenko1998alt-lower,
  title     = {{Lower Bounds for the Complexity of Learning Half-Spaces with Membership Queries}},
  author    = {Shevchenko, Valery N. and Zolotykh, Nikolai Yu.},
  booktitle = {International Conference on Algorithmic Learning Theory},
  year      = {1998},
  pages     = {61-71},
  doi       = {10.1007/3-540-49730-7_5},
  url       = {https://mlanthology.org/alt/1998/shevchenko1998alt-lower/}
}