融合异构特征的子空间迁移学习算法

张景祥; 王士同; 邓赵红; 蒋亦樟; 李奕

doi:10.3724/SP.J.1004.2014.00236

融合异构特征的子空间迁移学习算法

doi: 10.3724/SP.J.1004.2014.00236 cstr: 32138.14.SP.J.1004.2014.00236

1.
江南大学数字媒体学院无锡 214122;
2.
江南大学理学院无锡 214122

基金项目:

国家自然科学基金（61170122，61202311，61272210）;江苏省自然科学基金（BK2012552）资助

详细信息

作者简介:
王士同江南大学数字媒体学院教授．主要研究方向为人工智能，模式识别和生物信息.E-mail：wxwangst@yahoo.com.cn

计量
- 文章访问数: 3060
- HTML全文浏览量: 174
- PDF下载量: 1860
- 被引次数: 0
出版历程
- 收稿日期: 2012-12-31
- 修回日期: 2013-04-02
- 刊出日期: 2014-02-20

A Subspace Transfer Learning Algorithm Integrating Heterogeneous Features

1.
School of Digital Media, Jiangnan University, Wuxi 214122;
2.
School of Science, Jiangnan University, Wuxi 214122

Funds:

Supported by National Natural Science Foundation of China (61170122, 61202311, 61272210), and Natural Science Foundation of Jiangsu Province (BK2012552)

摘要

摘要: 特征迁移重在领域共有特征间学习，然而其忽略领域特有特征的判别信息，使算法的适应性受到一定的局限. 针对此问题，提出了一种融合异构特征的子空间迁移学习（The subspace transfer learning algorithm integrating with heterogeneous features，STL-IHF）算法.该算法将数据的特征空间看成共享和特有两个特征子空间的组合，同时基于经验风险最小框架将共享特征和特有特征共同嵌入到支持向量机（Support vector machine，SVM）的训练过程中.其在共享特征子空间上实现知识迁移的同时兼顾了领域特有的异构信息，增强了算法的适应性.模拟和真实数据集上的实验结果表明了所提方法的有效性.
- 特征空间 /
- 异构特征 /
- 迁移学习 /
- 分类 /
- 支持向量机
Abstract: The traditional transfer feature algorithms usually focus on learning by using the common features between the source domain and the target domain but ignore the discriminant information of the specific features of each domain, which makes the existing algorithms lack the adaptability to some extent. In order to circumvent this issue, in this paper a novel subspace transfer learning algorithm integrating heterogeneous features (STL-IHF) is proposed based on the empirical risk minimum framework. The proposed method is based on the support vector machine (SVM)-like framework with the feature space of each domain as a combination of the common features and the specified features. The proposed algorithm can not only realize the transfer learning from the common features but also effectively leverage the specified features of each domain, which makes it have much better adaptability in learning. Experimental results on simulation and real data set show the power of the proposed algorithm.
- Feature subspace /
- heterogeneous features /
- transfer learning /
- classification /
- support vector machine (SVM)

HTML全文

参考文献(28)

[1]	Pan J L, Yang Q. A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345-1359
[2]	Duan L X, Tsang I W, Xu D. Domains transfer multiple kernel learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(3): 465-479
[3]	Tu W T, Sun S L. A subject transfer framework for EEG classification. Neurocomputing, 2012, 82: 109-116
[4]	Daume Ⅲ H, Marcu D. Domain adaptation for statistical classifiers. Journal of Artificial Intelligence Research, 2006, 26(1): 101-126
[5]	Biekel S, Bruckner M, Scheffer T. Discriminative learning for differing training and test distributions. In: Proceedings of the 24th International Conference on Machine Learning. New York, USA: ACM, 2007. 81-88
[6]	Bickel S, Sawade C, Scheffer T. Transfer learning by distribution matching for targeted advertising. In: Proceedings of the 21st Annual Conference on Neural Information Processing Systems. Cambridge: MIT Press, 2009. 145-152
[7]	Wu P C, Dietterich T G. Improving SVM accuracy by training on auxiliary data sources. In: Proceedings of the 21st International Conference on Machine Learning (ICML). New York, USA: ACM, 2004. 110-117
[8]	Dai W Y, Yang Q, Xue G R, Yu Y. Boosting for transfer learning. In: Proceedings of the 24th International Conference on Machine Learning (ICML). New York, USA: ACM, 2007. 193-200
[9]	Quanz B, Huan J. Large margin transductive transfer learning. In: Proceedings of the 18th ACM Conference on Information and Knowledge Management (CIKM). New York, USA: ACM, 2009. 1327-1336
[10]	Xu Z J, Sun S L. Multi-view transfer learning with Adaboost. In: Proceedings of the 23rd Conference on Tools with Artificial Intelligence. Boca Raton, FL: IEEE, 2011. 399-402
[11]	Chen M M, Weinberger K Q, Blitzer J. Co-training for domain adaptation. In: Proceedings of the 25th Conference on Neural Information Processing Systems (NIPS). New York: USA: Curran Associates, Inc., 2011. 1231-1240
[12]	Xu Z J, Sun S L. Multi-source transfer learning with multi-view adaboost. Neural Information Processing, 2012, 7665: 332-339
[13]	Jiang Yi-Zhang, Deng Zhao-Hong, Wang Shi-Tong. Mamdani-Larsen type transfer learning fuzzy system. Acta Automatica Sinica, 2012, 38(9): 1393-1409 (蒋亦樟, 邓赵红, 王士同. ML型迁移学习模糊系统. 自动化学报, 2012, 38(9): 1393-1409)
[14]	Zhu Mei-Qiang, Cheng Yu-Hu, Li Ming, Wang Xue-Song, Feng Huan-Ting. A hybrid transfer algorithm for reinforcement learning based on spectral method. Acta Automatica Sinica, 2012, 38(11): 1765-1776 (朱美强, 程玉虎, 李明, 王雪松, 冯涣婷. 一类基于谱方法的强化学习混合迁移算法. 自动化学报, 2012, 38(11): 1765-1776)
[15]	Jiang W H, Chung F L. Transfer spectral clustering. In: Proceedings of the 2012 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD). Berlin, Heidelberg: Springer-Verlag, 2012. 789-803
[16]	Argyriou A, Micchelli C A, Pontil M, Ying Y M. A spectral regularization frame work for multi-task structure learning. In: Proceedings of Advances in Neural Information Processing Systems (NIPS 2008). Cambridge, MA: MIT Press, 2007. 25-32
[17]	Pan S J L, Kwok J T, Yang Q. Transfer learning via dimensionality reduction. In: Proceedings of the 23rd International Conference on Artificial Intelligence. California, USA: AAAI Press, 2008. 677-682
[18]	Pan S J L, Ni X C, Sun J T, Yang Q, Chen Z. Cross-domain sentiment classification via spectral feature alignment. In: Proceedings of the 19th International Conference on World Wide Web (WWW'10). New York, USA: ACM, 2010. 751-760
[19]	Tu W T, Sun S L. Transferable discriminative dimensionality reduction. In: Proceedings of the 23rd IEEE International Conference on Tools with Artificial Intelligence (CTAI). Boca Raton, FL: IEEE, 2011. 865-868
[20]	Gao X B, Wang X M, Li X L, Tao D C. Transfer latent variable model based on divergence analysis. Pattern Recognition, 2011, 44(10-11): 2358-2366
[21]	Gao X B, Wang Z, Yan P K, Li X L. Transfer learning for pedestrian detection. Neurocomputing, 2013, 100: 51-57
[22]	Gretton A, Fukumizu K, Harchaoui Z, Sriperumbudur B K. A fast, consistent kernel two-sample test. In: Proceedings of Advances in Neural Information Processing Systems (NIPS 2010). Red Hook, NY: MIT Press, 2010. 673-681
[23]	Gao Jun, Wang Shi-Tong, Deng Zhao-Hong. Global and local preserving based semi-supervised support vector machine. Acta Electronica Sinica, 2010, 38(7): 1626-1633 (皋军, 王士同, 邓赵红. 基于全局和局部保持的半监督支持向量机. 电子学报, 2010, 38(7): 1626-1633)
[24]	Deng Nai-Yang, Tian Ying-Jie. New Method in Data Mining: Support Vector Machine. Beijing: Science Press, 2004. 5-16 (邓乃阳, 田英杰. 数据挖掘中的新方法: 支持向量机. 北京: 科学出版社, 2004. 5-16)
[25]	Yuan Ya-Xiang. Optimization Theory and Methods. Beijing: Science Press, 1997. 176-189 (袁亚湘. 最优化理论与方法. 北京: 科学出版社, 1997. 176-189)
[26]	Tao Jian-Wen, Wang Shi-Tong. Kernel support vector machine for domain adaptation. Acta Automatica Sinica, 2012, 38(5): 797-881 (陶剑文, 王士同. 领域适应核支持向量机. 自动化学报, 2012, 38(5): 797-881)
[27]	Cai D, He X F, Han J W, Zhang H J. Orthogonal Laplacianfaces for face recognition. IEEE Transactions on Image Processing, 2006, 15(11): 3608-3614
[28]	Zhang J Y, Zhang B, Jiang X Z. Analyses of feature extraction methods based on wavelet transform. Signal Processing, 2000, 16(2): 156-162