ML Anthology

Weighted One-Against-All

Alina Beygelzimer, John Langford, Bianca Zadrozny
AAAI 2005 pp. 720-725
doi:10.5694/j.1326-5377.1972.tb108071.x /aaai/2005/beygelzimer2005aaai-weighted/

Abstract

The one-against-all reduction from multiclass classifi-cation to binary classification is a standard technique used to solve multiclass problems with binary classi-fiers. We show that modifying this technique in order to optimize its error transformation properties results in a superior technique, both experimentally and the-oretically. This algorithm can also be used to solve a more general classification problem “multi-label classi-fication, ” which is the same as multiclass classification except that it allows multiple correct labels for a given example.

PDF AAAI Semantic Scholar

Cite

Text

Beygelzimer et al. "Weighted One-Against-All." AAAI Conference on Artificial Intelligence, 2005. doi:10.5694/j.1326-5377.1972.tb108071.x

Markdown

[Beygelzimer et al. "Weighted One-Against-All." AAAI Conference on Artificial Intelligence, 2005.](https://mlanthology.org/aaai/2005/beygelzimer2005aaai-weighted/) doi:10.5694/j.1326-5377.1972.tb108071.x

BibTeX

@inproceedings{beygelzimer2005aaai-weighted,
  title     = {{Weighted One-Against-All}},
  author    = {Beygelzimer, Alina and Langford, John and Zadrozny, Bianca},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2005},
  pages     = {720-725},
  doi       = {10.5694/j.1326-5377.1972.tb108071.x},
  url       = {https://mlanthology.org/aaai/2005/beygelzimer2005aaai-weighted/}
}