Estimating Weighted Areas Under the ROC Curve

Abstract

Exponential bounds on the estimation error are given for the plug-in estimator of weighted areas under the ROC curve. The bounds hold for single score functions and uniformly over classes of functions, whose complexity can be controlled by Gaussian or Rademacher averages. The results justify learning algorithms which select score functions to maximize the empirical partial area under the curve (pAUC). They also illustrate the use of some recent advances in the theory of nonlinear empirical processes.

PDF NeurIPS Semantic Scholar

Cite

Text

Maurer and Pontil. "Estimating Weighted Areas Under the ROC Curve." Neural Information Processing Systems, 2020.

Markdown

[Maurer and Pontil. "Estimating Weighted Areas Under the ROC Curve." Neural Information Processing Systems, 2020.](https://mlanthology.org/neurips/2020/maurer2020neurips-estimating/)

BibTeX

@inproceedings{maurer2020neurips-estimating,
  title     = {{Estimating Weighted Areas Under the ROC Curve}},
  author    = {Maurer, Andreas and Pontil, Massimiliano},
  booktitle = {Neural Information Processing Systems},
  year      = {2020},
  url       = {https://mlanthology.org/neurips/2020/maurer2020neurips-estimating/}
}