论文信息 - Multi-class probabilistic classification using inductive and cross Venn-Abers predictors

Multi-class probabilistic classification using inductive and cross Venn-Abers predictors

Inductive (IVAP) and cross (CVAP) Venn–Abers predictors are computationally efficient algorithms for probabilistic prediction in binary classification problems. We present a new approach to multi-class probability estimation by turning IVAPs and CVAPs into multiclass probabilistic predictors. The proposed multi-class predictors are experimentally more accurate than both uncalibrated predictors and existing calibration methods.

Valery Manokhin | Valery Manokhin

[1] Rich Caruana,et al. Predicting good probabilities with supervised learning , 2005, ICML.

[2] Catherine Blake,et al. UCI Repository of machine learning databases , 1998 .

[3] Gérard Dreyfus,et al. Pairwise Neural Network Classifiers with Probabilistic Outputs , 1994, NIPS.

[4] Jason Weston,et al. Support vector machines for multi-class pattern recognition , 1999, ESANN.

[5] Leo Breiman,et al. Classification and Regression Trees , 1984 .

[6] Rich Caruana,et al. An empirical comparison of supervised learning algorithms , 2006, ICML.

[7] Vladimir Vovk,et al. Venn-Abers Predictors , 2012, UAI.

[8] David J. Spiegelhalter,et al. Machine Learning, Neural and Statistical Classification , 2009 .

[9] Yoshua Bengio,et al. Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[10] John Platt,et al. Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[11] Bianca Zadrozny,et al. Obtaining calibrated probability estimates from decision trees and naive Bayesian classifiers , 2001, ICML.

[12] Vladimir Vovk,et al. Large-scale probabilistic predictors with and without guarantees of validity , 2015, NIPS.