PAC-Bayesian Theory for Transductive Learning

被引：0

作者：

Begin, Luc ^{[1
]}

Germain, Pascal ^{[2
]}

Laviolette, Francois ^{[2
]}

Roy, Jean-Francis ^{[2
]}

机构：

[1] Univ Moncton, Campus Edmundston, Moncton, NB, Canada

[2] Univ Laval, Dept Informat & Genie Logiciel, Quebec City, PQ, Canada

来源：

ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33 | 2014年 / 33卷

基金：

加拿大自然科学与工程研究理事会;

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We propose a PAC-Bayesian analysis of the transductive learning setting, introduced by Vapnik [1998], by proposing a family of new bounds on the generalization error. Some of them are derived from their counterpart in the inductive setting, and others are new. We also compare their behavior.

引用

页码：105 / 113

页数：9

共 15 条

[1]

[Anonymous], CORR

[2]

Blake C. L., 1998, Uci repository of machine learning databases

[3] Explicit learning curves for transduction and application to clustering and compression algorithms [J].