An efficient model-free estimation of multiclass conditional probability

DOI10.1016/J.JSPI.2013.08.008MaRDI QIDQ393624zbMATH OpenOpenAlexFDO

Authors Tu Xu, Junhui Wang

Publication date 23 January 2014

Published in Journal of Statistical Planning and Inference (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1209.4951

zbMATH Keywords

quantile regressionprobability estimationtuninginterval estimationmulticlass classification

Mathematics Subject Classification ID

Density estimation (62G07) Nonparametric regression and quantile regression (62G08)

Abstract: Conventional multiclass conditional probability estimation methods, such as Fisher's discriminate analysis and logistic regression, often require restrictive distributional model assumption. In this paper, a model-free estimation method is proposed to estimate multiclass conditional probability through a series of conditional quantile regression functions. Specifically, the conditional class probability is formulated as difference of corresponding cumulative distribution functions, where the cumulative distribution functions can be converted from the estimated conditional quantile regression functions. The proposed estimation method is also efficient as its computation cost does not increase exponentially with the number of classes. The theoretical and numerical studies demonstrate that the proposed estimation method is highly competitive against the existing competitors, especially when the number of classes is relatively large.

Recommendations

• Robust Model-Free Multiclass Probability Estimation
• Multiclass Boosting Algorithms for Shrinkage Estimators of Class Probability
• scientific article; zbMATH DE number 1931829
• Estimating conditional probabilities by mixtures of low order conditional distributions
• Boosting conditional probability estimators

Cites work

• scientific article; zbMATH DE number 45848 (Why is no real title available?)
• Bayesian empirical likelihood for quantile regression
• Bi-level path following for cross validated solution of kernel quantile regression
• Bivariate Quantile Smoothing Splines
• Classification by pairwise coupling
• Estimation of generalization error: random and fixed inputs
• Hard or soft classification? Large-margin unified machines
• Noncrossing quantile regression curve estimation
• Nonparametric Conditional Density Estimation Using Piecewise-Linear Solution Path of Kernel Quantile Regression
• On the influence of the kernel on the consistency of support vector machines
• Optimal Model Assessment, Selection, and Combination
• Probability estimates for multi-class classification by pairwise coupling
• Probability estimation for large-margin classifiers
• Quantile Regression in Reproducing Kernel Hilbert Spaces
• Quantile estimation for discrete data via empirical likelihood
• Quantile regression in partially linear varying coefficient models
• Quantile regression.
• Quantiles for Counts
• Regression Quantiles
• Robust Model-Free Multiclass Probability Estimation
• Simultaneous multiple non-crossing quantile regression estimation using kernel constraints
• Soft and hard classification by reproducing kernel Hilbert space methods
• Stepwise multiple quantile regression estimation using non-crossing constraints
• The Estimation of Prediction Error
• The Little Bootstrap and Other Methods for Dimensionality Selection in Regression: X-Fixed Prediction Error
• The covering number in learning theory

Cited in

(4)

• Multiclass Probability Estimation With Support Vector Machines
• Robust Model-Free Multiclass Probability Estimation
• Multiclass classification of the scalar Gaussian random field observation with known spatial correlation function
• Probability estimation with machine learning methods for dichotomous and multicategory outcome: theory

This page was built for publication: An efficient model-free estimation of multiclass conditional probability

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q393624)