2000 character limit reached
On Classification-Calibration of Gamma-Phi Losses (2302.07321v2)
Published 14 Feb 2023 in stat.ML and cs.LG
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.
- Robust bi-tempered logistic loss based on Bregman divergences. Advances in Neural Information Processing Systems, 32, 2019.
- Calibration and consistency of adversarial surrogate losses. Advances in Neural Information Processing Systems, 34:9804–9815, 2021.
- Multi-Class H𝐻Hitalic_H-Consistency Bounds. In Advances in Neural Information Processing Systems, 2022.
- AdaBoost is consistent. Advances in Neural Information Processing Systems, 19, 2006.
- Convexity, classification, and risk bounds. Journal of the American Statistical Association, 101(473):138–156, 2006.
- Guess-averse loss functions for cost-sensitive multiclass boosting. In International Conference on Machine Learning, pages 586–594. PMLR, 2014.
- Geometric deep learning: Grids, groups, graphs, geodesics, and gauges. arXiv preprint arXiv:2104.13478, 2021.
- Multiclass classification, information, divergence and surrogate risk. The Annals of Statistics, 46(6B):3246–3275, 2018.
- Convex analysis at infinity: An introduction to astral space. arXiv preprint arXiv:2205.03260, 2022.
- An embedding framework for consistent polyhedral surrogates. Advances in Neural Information Processing Systems, 32, 2019.
- Peter J Huber. Robust statistics. In International Encyclopedia of Statistical Science, pages 1248–1251. Springer, 2011.
- Online multiclass boosting. Advances in Neural Information Processing Systems, 30, 2017.
- Consistency versus realizable h-consistency for multiclass classification. In International Conference on Machine Learning, pages 801–809. PMLR, 2013.
- On the design of loss functions for classification: theory, robustness to outliers, and savageboost. Advances in Neural Information Processing Systems, 21, 2008.
- A theory of multiclass boosting. Journal of Machine Learning Research, 2013.
- Applied functional analysis. Chapman and Hall/CRC, 2017.
- Multiclass classification calibration functions. arXiv preprint arXiv:1609.06385, 2016.
- Online agnostic multiclass boosting. In Advances in Neural Information Processing Systems, 2022.
- Convex calibration dimension for multiclass loss matrices. The Journal of Machine Learning Research, 17(1):397–441, 2016.
- Consistent algorithms for multiclass classification with an abstain option. Electronic Journal of Statistics, 12(1):530 – 554, 2018.
- Multiclass boosting: Theory and algorithms. Advances in Neural Information Processing Systems, 24, 2011.
- Multiclass boosting: Margins, codewords, losses, and algorithms. Journal of Machine Learning Research, 20(137):1–68, 2019.
- Ingo Steinwart. How to compare different loss functions and their risks. Constructive Approximation, 26(2):225–287, 2007.
- On the consistency of multiclass classification methods. Journal of Machine Learning Research, 8(5), 2007.
- Weston-Watkins hinge loss and ordered partitions. Advances in Neural Information Processing Systems, 33:19873–19883, 2020.
- Bayes consistency vs. H𝐻Hitalic_H-consistency: The interplay between surrogate loss functions and the scoring function class. Advances in Neural Information Processing Systems, 33:16927–16936, 2020.
- Tong Zhang. Statistical analysis of some multi-category large margin classification methods. Journal of Machine Learning Research, 5(Oct):1225–1251, 2004.
- Coherence functions for multicategory margin-based classification methods. In Artificial Intelligence and Statistics, pages 647–654. PMLR, 2009.