Surrogate Losses and Regret Bounds for Cost-Sensitive Classification with Example-Dependent Costs
Abstract
We study surrogate losses in the context of cost-sensitive classification with example-dependent costs, a problem also known as regression level set estimation. We give sufficient conditions on the surrogate loss for the existence of a surrogate regret bound. Such bounds imply that as the surrogate risk tends to its optimal value, so too does the expected misclassification cost. Our sufficient conditions encompass example-dependent versions of the hinge, exponential, and other common losses. These results provide theoretical justification for some previously proposed surrogate-based algorithms, and suggests others that have not yet been developed.
Cite
Text
Scott. "Surrogate Losses and Regret Bounds for Cost-Sensitive Classification with Example-Dependent Costs." International Conference on Machine Learning, 2011.Markdown
[Scott. "Surrogate Losses and Regret Bounds for Cost-Sensitive Classification with Example-Dependent Costs." International Conference on Machine Learning, 2011.](https://mlanthology.org/icml/2011/scott2011icml-surrogate/)BibTeX
@inproceedings{scott2011icml-surrogate,
title = {{Surrogate Losses and Regret Bounds for Cost-Sensitive Classification with Example-Dependent Costs}},
author = {Scott, Clayton},
booktitle = {International Conference on Machine Learning},
year = {2011},
pages = {153-160},
url = {https://mlanthology.org/icml/2011/scott2011icml-surrogate/}
}