ML Anthology

Robust Reductions from Ranking to Classification

Maria-Florina Balcan, Nikhil Bansal, Alina Beygelzimer, Don Coppersmith, John Langford, Gregory B. Sorkin
MLJ 2008 pp. 139-153
doi:10.1007/S10994-008-5058-6 /mlj/2008/balcan2008mlj-robust/

Abstract

We reduce ranking, as measured by the Area Under the Receiver Operating Characteristic Curve (AUC), to binary classification. The core theorem shows that a binary classification regret of r on the induced binary problem implies an AUC regret of at most 2 r . This is a large improvement over approaches such as ordering according to regressed scores, which have a regret transform of r ↦ nr where n is the number of elements.

PDF MLJ Semantic Scholar

Cite

Text

Balcan et al. "Robust Reductions from Ranking to Classification." Machine Learning, 2008. doi:10.1007/S10994-008-5058-6

Markdown

[Balcan et al. "Robust Reductions from Ranking to Classification." Machine Learning, 2008.](https://mlanthology.org/mlj/2008/balcan2008mlj-robust/) doi:10.1007/S10994-008-5058-6

BibTeX

@article{balcan2008mlj-robust,
  title     = {{Robust Reductions from Ranking to Classification}},
  author    = {Balcan, Maria-Florina and Bansal, Nikhil and Beygelzimer, Alina and Coppersmith, Don and Langford, John and Sorkin, Gregory B.},
  journal   = {Machine Learning},
  year      = {2008},
  pages     = {139-153},
  doi       = {10.1007/S10994-008-5058-6},
  volume    = {72},
  url       = {https://mlanthology.org/mlj/2008/balcan2008mlj-robust/}
}