On Classification-Calibration of Gamma-Phi Losses
Yutong Wang, Clayton D. Scott

TL;DR
This paper establishes a general sufficient condition for the classification-calibration of Gamma-Phi losses, a family of multiclass loss functions, providing the first fully justified nonconvex surrogate losses for multiclass classification.
Contribution
It introduces the first general sufficient condition for classification-calibration of Gamma-Phi losses and corrects a misconception about a previously proposed condition.
Findings
First general sufficient condition for CC of Gamma-Phi losses
Identification of a previously proposed insufficient condition
Highlights a technical issue in multiclass CC analysis
Abstract
Gamma-Phi losses constitute a family of multiclass classification loss functions that generalize the logistic and other common losses, and have found application in the boosting literature. We establish the first general sufficient condition for the classification-calibration (CC) of such losses. To our knowledge, this sufficient condition gives the first family of nonconvex multiclass surrogate losses for which CC has been fully justified. In addition, we show that a previously proposed sufficient condition is in fact not sufficient. This contribution highlights a technical issue that is important in the study of multiclass CC but has been neglected in prior work.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsBlind Source Separation Techniques · Wireless Signal Modulation Classification · Fault Detection and Control Systems
