PAC-Bayesian high dimensional bipartite ranking

被引：0

作者：

Guedj, Benjamin ^{[1
]}

Robbiano, Sylvain ^{[2
]}

机构：

[1] INRIA, Modal Project Team, Rennes, France

[2] UCL, Dept Stat Sci, London, England

来源：

JOURNAL OF STATISTICAL PLANNING AND INFERENCE | 2018年 / 196卷

关键词：

Bipartite ranking; High dimension and sparsity; MCMC; PAC-Bayesian aggregation; Supervised statistical learning; BOUNDS; MINIMIZATION; AGGREGATION; REGRESSION; MODEL;

D O I：

10.1016/j.jspi.2017.10.010

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

This paper is devoted to the bipartite ranking problem, a classical statistical learning task, in a high dimensional setting. We propose a scoring and ranking strategy based on the PAC-Bayesian approach. We consider nonlinear additive scoring functions, and we derive non-asymptotic risk bounds under a sparsity assumption. In particular, oracle inequalities in probability holding under a margin condition assess the performance of our procedure, and prove its minimax optimality. An MCMC-flavored algorithm is proposed to implement our method, along with its behavior on synthetic and real-life datasets. (C) 2017 Elsevier B.V. All rights reserved.

引用

页码：70 / 86

页数：17

共 50 条

[1] PAC-Bayesian Inequalities for Martingales
Seldin, Yevgeny
Laviolette, Francois
Cesa-Bianchi, Nicolo
Shawe-Taylor, John
Auer, Peter
IEEE TRANSACTIONS ON INFORMATION THEORY, 2012, 58 (12) : 7086 - 7093
[2] Some PAC-Bayesian Theorems
David A. McAllester
Machine Learning, 1999, 37 : 355 - 363
[3] PAC-Bayesian generic chaining
Audibert, JY
Bousquet, O
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 16, 2004, 16 : 1125 - 1132
[4] PAC-Bayesian Collective Stability
London, Ben
Huang, Bert
Taskar, Ben
Getoor, Lise
ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33, 2014, 33 : 585 - 594
[5] Some PAC-Bayesian theorems
McAllester, DA
MACHINE LEARNING, 1999, 37 (03) : 355 - 363
[6] PAC-Bayesian Theory Meets Bayesian Inference
Germain, Pascal
Bach, Francis
Lacoste, Alexandre
Lacoste-Julien, Simon
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 29 (NIPS 2016), 2016, 29
[7] PAC-Bayesian Theory for Transductive Learning
Begin, Luc
Germain, Pascal
Laviolette, Francois
Roy, Jean-Francis
ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33, 2014, 33 : 105 - 113
[8] PAC-Bayesian stochastic model selection
McAllester, DA
MACHINE LEARNING, 2003, 51 (01) : 5 - 21
[9] A PAC-Bayesian Bound for Lifelong Learning
Pentina, Anastasia
Lampert, Christoph H.
INTERNATIONAL CONFERENCE ON MACHINE LEARNING, VOL 32 (CYCLE 2), 2014, 32 : 991 - 999
[10] PAC-Bayesian AUC classification and scoring
Ridgway, James
Alquier, Pierre
Chopin, Nicolas
Liang, Feng
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 27 (NIPS 2014), 2014, 27

← 1 2 3 4 5 →