From Cost-Sensitive to Tight F-measure Bounds

Kevin Bascol, Rémi Emonet, Elisa Fromont, Amaury Habrard, Guillaume Metzler, Marc Sebban
Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, PMLR 89:1245-1253, 2019.

Abstract

The F-measure is a classification performance measure, especially suited when dealing with imbalanced datasets, which provides a compromise between the precision and the recall of a classifier. As this measure is non convex and non linear, it is often indirectly optimized using cost-sensitive learning (that affects different costs to false positives and false negatives). In this article, we derive theoretical guarantees that give tight bounds on the best F-measure that can be obtained from cost-sensitive learning. We also give an original geometric interpretation of the bounds that serves as an inspiration for CONE, a new algorithm to optimize for the F-measure. Using 10 datasets exhibiting varied class imbalance, we illustrate that our bounds are much tighter than previous work and show that CONE learns models with either superior F-measures than existing methods or comparable but in fewer iterations.

Cite this Paper

BibTeX
@InProceedings{pmlr-v89-bascol19a, title = {From Cost-Sensitive to Tight F-measure Bounds}, author = {Bascol, Kevin and Emonet, R\'{e}mi and Fromont, Elisa and Habrard, Amaury and Metzler, Guillaume and Sebban, Marc}, booktitle = {Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics}, pages = {1245--1253}, year = {2019}, editor = {Chaudhuri, Kamalika and Sugiyama, Masashi}, volume = {89}, series = {Proceedings of Machine Learning Research}, month = {16--18 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v89/bascol19a/bascol19a.pdf}, url = {https://proceedings.mlr.press/v89/bascol19a.html}, abstract = {The F-measure is a classification performance measure, especially suited when dealing with imbalanced datasets, which provides a compromise between the precision and the recall of a classifier. As this measure is non convex and non linear, it is often indirectly optimized using cost-sensitive learning (that affects different costs to false positives and false negatives). In this article, we derive theoretical guarantees that give tight bounds on the best F-measure that can be obtained from cost-sensitive learning. We also give an original geometric interpretation of the bounds that serves as an inspiration for CONE, a new algorithm to optimize for the F-measure. Using 10 datasets exhibiting varied class imbalance, we illustrate that our bounds are much tighter than previous work and show that CONE learns models with either superior F-measures than existing methods or comparable but in fewer iterations.} }
Endnote
%0 Conference Paper %T From Cost-Sensitive to Tight F-measure Bounds %A Kevin Bascol %A Rémi Emonet %A Elisa Fromont %A Amaury Habrard %A Guillaume Metzler %A Marc Sebban %B Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Masashi Sugiyama %F pmlr-v89-bascol19a %I PMLR %P 1245--1253 %U https://proceedings.mlr.press/v89/bascol19a.html %V 89 %X The F-measure is a classification performance measure, especially suited when dealing with imbalanced datasets, which provides a compromise between the precision and the recall of a classifier. As this measure is non convex and non linear, it is often indirectly optimized using cost-sensitive learning (that affects different costs to false positives and false negatives). In this article, we derive theoretical guarantees that give tight bounds on the best F-measure that can be obtained from cost-sensitive learning. We also give an original geometric interpretation of the bounds that serves as an inspiration for CONE, a new algorithm to optimize for the F-measure. Using 10 datasets exhibiting varied class imbalance, we illustrate that our bounds are much tighter than previous work and show that CONE learns models with either superior F-measures than existing methods or comparable but in fewer iterations.
APA
Bascol, K., Emonet, R., Fromont, E., Habrard, A., Metzler, G. & Sebban, M.. (2019). From Cost-Sensitive to Tight F-measure Bounds. Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 89:1245-1253 Available from https://proceedings.mlr.press/v89/bascol19a.html.

Related Material