Batch and Online Learning Algorithms for Nonconvex Neyman-Pearson Classification

被引：21

作者：

Gasso, Gilles ^{[1
]}

Pappaioannou, Aristidis ^{[2
]}

Spivak, Marina ^{[3
]}

Bottou, Leon ^{[4
]}

机构：

[1] INSA Rouen, LITIS, Rouen, France

[2] Ecole Polytech Fed Lausanne, CH-1015 Lausanne, Switzerland

[3] NYU, New York, NY USA

[4] NEC Labs, Princeton, NJ USA

来源：

ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY | 2011年 / 2卷 / 03期

关键词：

Algorithms; Neyman-Pearson; nonconvex SVM; DC algorithm; online learning; PEPTIDE IDENTIFICATION; OPTIMIZATION;

D O I：

10.1145/1961189.1961200

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We describe and evaluate two algorithms for Neyman-Pearson (NP) classification problem which has been recently shown to be of a particular importance for bipartite ranking problems. NP classification is a nonconvex problem involving a constraint on false negatives rate. We investigated batch algorithm based on DC programming and stochastic gradient method well suited for large-scale datasets. Empirical evidences illustrate the potential of the proposed methods.

引用

页数：19

共 29 条

[1]

ANDRIEU L, 2007, STOCHASTIC PROGRAMMI

[2]

[Anonymous], ADV NEURAL INFORM PR

[3]

[Anonymous], 1958, STUD LINEAR NONLINEA

[4]

[Anonymous], P INT C MACH LEARN

[5]

[Anonymous], 2006, Proceedings of the 23rd International Conference on Machine Learning

[6]

[Anonymous], 1971, Adaptation and Learning in Automatic Systems

[7]

Bach FR, 2006, J MACH LEARN RES, V7, P1713

[8]

Bottou L., 2007, C ADV NEUR INF PROC

[9] General solution and learning method for binary classification with performance constraints [J].

Bounsiar, Abdenour ;

Beauseroy, Pierre ;

Grall-Maes, Edith .

PATTERN RECOGNITION LETTERS, 2008, 29 (10) :1455-1465

[10]

Ciarlet P.G., 1989, INTRO NUMERICAL LINE, DOI DOI 10.1017/9781139171984

← 1 2 3 →