New multicategory boosting algorithms based on multicategory Fisher-consistent losses

DOI10.1214/08-AOAS198zbMath1158.62044arXiv0901.3988WikidataQ42081059 ScholiaQ42081059MaRDI QIDQ999662

Ji Zhu, Hui Zou, Trevor Hastie

Publication date: 10 February 2009

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/0901.3988

zbMATH Keywords

boostingmulticategory classificationFisher-consistent losses

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Related Items

Angle-based cost-sensitive multicategory classification, Multi-class boosting with asymmetric binary weak-learners, A Fisher consistent multiclass loss function with variable margin on positive examples, Group collaborative representation for image set classification, Adaptively weighted large-margin angle-based classifiers, Multicategory vertex discriminant analysis for high-dimensional data, Deformation of log-likelihood loss function for multiclass boosting, Multicategory large margin classification methods: hinge losses vs. coherence functions, An improved multiclass LogitBoost using adaptive-one-vs-one, Multiclass Probability Estimation With Support Vector Machines, On the Effect and Remedies of Shrinkage on Classification Probability Estimation

Uses Software

• UCI-ml
• ElemStatLearn

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Greedy function approximation: A gradient boosting machine.
• Boosting algorithms: regularization, prediction and model fitting
• A decision-theoretic generalization of on-line learning and an application to boosting
• A note on margin-based loss functions in classification
• Additive logistic regression: a statistical view of boosting. (With discussion and a rejoinder by the authors)
• Empirical margin distributions and bounding the generalization error of combined classifiers
• On the Bayes-risk consistency of regularized boosting methods.
• Statistical behavior and consistency of classification methods based on convex risk minimization.
• Improved boosting algorithms using confidence-rated predictions
• 10.1162/15324430152733133
• Boosting With theL₂Loss
• 10.1162/1532443041424319
• Multicategory Support Vector Machines
• Multicategory ψ-Learning
• The elements of statistical learning. Data mining, inference, and prediction