《机电工程》杂志,月刊( 详细... )

中国标准连续出版物号 ISSN 1001-4551 CN 33-1088/TH
主办单位浙江省机电集团有限公司
浙江大学
主编赵 群
副 主 编唐任仲、罗向阳(执行主编)
总 经 理罗向阳
出 版浙江《机电工程》杂志社有限公司
地 址杭州市上城区延安路95号浙江省机电集团大楼二楼211、212室
电话Tel+86-571-87041360、87239525
E-mailmeem_contribute@163.com
国外发行中国国际图书贸易总公司
订阅全国各地邮局   国外代号M3135
国内发行浙江省报刊发行局
邮发代号32-68
广告发布登记证:杭上市管广发G-001号

在线杂志

当前位置: 机电工程 >>在线杂志

基于小样本的集成学习研究

作者:倪勇1,2,吴汶芪2,李君2 日期:2009-12-30/span> 浏览:2880 查看PDF文档

基于小样本的集成学习研究

倪勇1,2,吴汶芪2,李君2
(1.浙江工业大学 信息学院,浙江 杭州 310004;2.浙江机电职业技术学院 电子信息工程系,浙江 杭州 310053)

摘要:为了提高集成学习在小数据量的有标记样本问题中的性能,结合半监督学习和选择性集成学习,提出了一种基于半监督回归的选择性集成算法SSRES。一方面,充分利用大量的未标记样本来辅助有标记样本的学习;另一方面,使用选择性集成学习进一步提高学习系统的泛化能力。实验结果表明,SSRES算法能够利用未标记样本和选择性集成技术提高学习器的性能。
关键词:集成学习;选择性集成;半监督学习;选择性集成算法
中图分类号:TP181文献标识码:A文章编号:1001-4551(2009)12-0041-04

Research of ensemble learning based on small samples
NI Yong1,2, WU Wen-qi2, LI Jun2
(1.College of Information Engineering, Zhejiang University of Technology, Hangzhou 310004, China; 2.Department of
Electronic & Information Engineering, Zhejiang Institute of Mechanical & Electric Engineering, Hangzhou 310053, China)
Abstract: In order to improve the performance of ensemble learning in a few labeled training examples, combining with semi-supervised learning and selective integration of learning, a new selective integration of algorithm based on semi-supervised regression namely SSRES was proposed. On the one hand, the method of a large number of unlabeled examples was used to reduce the requirement of labeled examples. On the other hand, selective integration learn was used to further improve the generalization ability of learning systems. Experiment results show that SSRES algorithm can improve the performance of learners with unlabeled example and selective integration of technology.
Key words: ensemble learning; selective integration; semi-supervised learning; selective integration of algorithm
参考文献(References):
[1]DIETTERICH T G. Machine-learning research: four current directions[J]. AI Magazine,1997,18 (4):97-136.
[2]ZHU X. Semi-supervised learning literature survey[R]. Technical Report 1530, Department of Computer, University of Wisconsin at Madison, Madison, WI, Apr,2006.
[3]周志华,王珏.机器学习及其应用2006[M].北京:清华大学出版社,2006.
[4]ZHOU Z H, WU J, TANG W. Ensembling neural networks: many could be better than all[J].Artificial Intelligence,2002,137(1-2):239-263.
[5]TSOUMAKAS G, PARTALAS I, VLAHAVAS I. A Taxonomy and Short Review of Ensemble Selection[C]. ECAI, Workshop on Supervised and Unsupervised Ensemble Methods and Their Applications, Patras, Greece,2008.
[6]HERNANDEZ-LOBATO D, MARTINEZ-MUNOZ G, SUAREZ A. Pruning in ordered regression bagging ensembles[C]//Proceedings of the IEEE World Congress on Computational Intelligence. Vancouver: [s.n.],2006:1266-1273.
[7]CHAPELLE O, ZIE A, SCHZLKOPF B. Semi-Supervised Learning[M]. MIT Press,2006.
[8]ZHOU Z H, CHEN K J, DAI H B. Enhancing relevance feedback in image retrieval using unlabeled data[J]. ACM Transactions on Information Systems,2006,24(2):219-244.
[9]ZHOU H Z, LI M. Semi-supervised learning with co-training[C]//Proceedings of the 19th International Joint Conference on Artificial Intelligence (IJCAI05). Edinburgh: [s.n.],2005:908-91.



友情链接

浙江机械信息网