Heterogeneous transfer learning based on translation invariant kernels

doi:10.7523/j.issn.2095-6134.2015.01.020

Abstract

Abstract:

We propose a new heterogeneous transfer learning method, which uses related heterogeneous feature dataset. We use translation invariant kernels(Euclidean kernels and RBF kernels) to map the target dataset and the related dataset to a new reproducing kernel Hilbert space, in which the two datasets have equal feature dimensions and similar distributions and reserve their topological property. The experimental results show that our method works well and the method based on the Euclidean kernel improves accuracy by more than 5%~10%.

Key words: heterogeneous transfer learning, translation invariant kernel, RKHS

CLC Number:

TP301

GUAN Zengda, CHENG Li, ZHU Tingshao. Heterogeneous transfer learning based on translation invariant kernels[J]. , 2015, 32(1): 121-126.

References

[1] Pan S J, Yang Q. A survey on transfer learning[J].IEEE Transactions on Knowledge and Data Engineering , 2010,22(10):1 345-1 359.

[2] Yang Q, Chen Y, Xue G, et al. Heterogeneous transfer learning for image clustering via the social web[C]//Proceedings of the 47th Annual Meeting of the ACL and the 4th IJCNLP of the AFNLP. Suntec, Singapore, 2009:1-9.

[3] Zhu Y, Chen Y, Lu Z, et al. Heterogeneous transfer learning for image classification[C]//Proceedings of the 25th AAAI Conference on Artificial Intelligence. San Francisco,USA, 2011:1 304-1 309.

[4] Dai W Y, Chen Y Q, Xue G R, et al. Translated learning: transfer learning across different feature spaces[C]//Proc 21st Ann Conf Neural Information Processing Systems. Vancouver, Canada, 2008.

[5] Wang C, Mahadevan S. Heterogeneous domain adaptation using manifold alignment[C]//Proc 22nd International Joint Conference on Artificial Intelligence. Barcelona, Spain, 2011: 1 541-1 546.

[6] Duan L, Xu D, Tsang I W. Learning with augmented features for heterogeneous domain adaptation[C]//Proceedings of the 29th International Conference on Machine Learning. Edinburgh, Scotland, UK, 2012.

[7] Kulis B, Saenko K, Darrell T. What you saw is not what you get: domain adaptation using asymmetric kernel transforms[C]//Computer Vision and Pattern Recognition, Colorado. USA, 2011: 1 785-1 792.

[8] Guan Z D, Bai S B, Zhu T S. Heterogeneous domain adaptation using linear kernel[C]//ICPCA-SWS. Vina del Mar, Chile, Springer, 2013.

[9] Gu Q Q, Li Z H, Han J W. Learning a kernel for multi-task clustering[C]//Proceedings of the 25th AAAI Conference on Artificial Intelligence. San Francisco, USA, 2011.

[10] Pan S J, Tsang I W, Kwok J T, et al. Domain adaptation via transfer component analysis[J]. IEEE Transactions on Neural Networks, 2011, 22(2):199-210.

[11] Gretton A, Borgwardt K M, Rasch M J, et al. A kernel method for the two-sample-problem[C]//Proceedings of the 2006 Conference Advances in Neural Information Processing Systems 19. Vancouver, Canada, MIT Press, 2006, 20:513-520.

[1]	WANG Jing, CHEN Lan, ZHANG He, WANG Haiyong. Multi-resource scheduling algorithm based on EDA simulation software [J]. Journal of University of Chinese Academy of Sciences, 2021, 38(5): 696-701.
[2]	LI Xiaomu, TIAN Miaomiao, MA Guangbin, LIN Youming, CHENG Bo. Satellite imaging planning method based on improved cuckoo search [J]. Journal of University of Chinese Academy of Sciences, 0, (): 544-544.
[3]	ZHANG Libo, ZHANG Fei, LUO Tiejian. Knowledge structure evolution and evaluation method in computer science [J]. , 2016, 33(6): 844-850.
[4]	QIAN Quan, ZHU Wei, CHE Hong-Yi, ZHANG Rui. An improved multi-pattern string matching algorithm and GPU parallelization [J]. Journal of University of Chinese Academy of Sciences, 2013, 30(5): 706-712.
[5]	GUO Long-Kun, SHEN Hong. Hardness of the edge-disjoint Min-Min problem in undirected graphs [J]. , 2012, 29(4): 549-554.
[6]	LI Hong-Yan, ZHAO Xian-Feng, HUANG Wei, SHENG Ren-Nong. Steganalysis of palette images based on run-length statistics and Walsh spectrum energy distribution [J]. , 2012, (3): 423-428.
[7]	ZHANG Tao, ZHAO Xian-Feng, HUANG Wei, SHENG Ren-Nong. Analysis of complementarity among three sets of the features and their fusion for spatial steganography [J]. , 2012, (2): 264-270.
[8]	WAN Wei, ZHAO Xian-Feng, HUANG Wei, SHENG Ren-Nong. Steganalysis of MP3Stego based on Huffman table distribution and recoding [J]. , 2012, 29(1): 118-124.
[9]	SHAO Xiu-Feng, LI Rong-Hua. On the security proof of a protocol for private integer comparison [J]. , 2011, 28(2): 262-265.
[10]	DENG Yi, ZHAO Xian-Feng, FENG Deng-Guo. JPEG image steganalysis based on characteristic function and quality factor [J]. , 2010, 27(3): 389-396.
[11]	WU Jie, CHEN Yu-Fu. Lifting algorithms for Gröbner basis computation of invariant ideals [J]. , 2009, 26(6): 731-744.
[12]	WANG Jing, XIA Lu-Ning, JING Ji-Wu, GU Qing. A fault-tolerant dedicated computer system model [J]. , 2009, 26(6): 826-834.
[13]	YIN Yin, LI Bao. Cryptanalysis and Improvement of Client-to-Client Password Authenticated Key Exchange Protocol [J]. Journal of University of Chinese Academy of Sciences, 2006, 23(6): 793-801.
[14]	ZHU Yu, ZHANG Hong-Bin. Selection Method for AVL Tree Rebalancing [J]. , 2006, 23(4): 527-533.
[15]	Tong Jianjun, Zou Mingfu. Detection Algorithm of Hough Transform in the Container Number Recognition [J]. , 2005, 22(1): 72-77.