© The Institution of Engineering and Technology
Domain adaptation is used for machine learning tasks, when the distribution of the training (obtained from source domain) set differs from that of the testing (referred as target domain) set. In the work presented in this study, the problem of unsupervised domain adaptation is solved using a novel optimisation function to minimise the global and local discrepancies between the transformed source and the target domains. The dissimilarity in data distributions is the major contributor to the global discrepancy between the two domains. The authors propose two techniques to preserve the local structural information of source domain: (i) identify closest pair of instances in source domain and minimise the distances between these pairs of instances after transformation; (ii) preserve the naturally occurring clusters present in source domain during transformation. This cost function and constraints yield a non-linear optimisation problem, used to estimate the weight matrix. An iterative framework solves the optimisation problem, providing a sub-optimal solution. Next, using orthogonality constraint, an optimisation task is formulated in the Stiefel manifold. Performance analysis using real-world datasets show that the proposed methods perform better than a few recently published state-of-the-art methods.
References
-
-
1)
-
22. Gong, B., Shi, Y., Sha, F., et al: ‘Geodesic flow kernel for unsupervised domain adaptation’. IEEE Conf. on Computer Vision and Pattern Recognition, 2012.
-
2)
-
3. Saenko, K., Kulis, B., Fritz, M., et al: ‘Adapting visual category models to new domains’. European Conf. on Computer vision, 2010.
-
3)
-
28. Wan, C., Pan, R., Li, J.: ‘Bi-weighting domain adaptation for cross-language text classification’. Int. Joint Conf. on Artificial Intelligence, 2011.
-
4)
-
21. Jhuo, I.H., Liu, D., Lee, D.T., et al: ‘Robust visual domain adaptation with low-rank reconstruction’. IEEE Conf. on Computer Vision and Pattern Recognition, 2012.
-
5)
-
32. Tagare, H.D.: ‘Notes on optimization on Stiefel manifolds’, , Yale University, 2011.
-
6)
-
16. Sugiyama, M., Nakajima, S., Kashima, H., et al: ‘Direct importance estimation with model selection and its application to covariate shift adaptation’. Proc. of Neural Information Processing Systems, 2007.
-
7)
-
17. Gopalan, R., Li, R., Chellappa, R.: ‘Domain adaptation for object recognition: an unsupervised approach’. Int. Conf. in Computer Vision, 2011.
-
8)
-
26. Baktashmotlagh, M., Harandi, M.T., Lovell, B.C., et al: ‘Unsupervised domain adaptation by domain invariant projection’. Int. Conf. on Computer Vision, 2013.
-
9)
-
9. Aytar, Y., Zisserman, A.: ‘Tabula rasa: model transfer for object category detection’. Int. Conf. on Computer Vision, 2011.
-
10)
-
24. Duan, L., Xu, D., Tsang, I.W., et al: ‘Visual event recognition in videos by learning from web data’, IEEE Trans. Pattern Anal. Mach. Intell., 2012, 34, (9), pp. 1667–1680.
-
11)
-
12)
-
7. Jiang, W., Zavesky, E., Chang, S., et al: ‘Cross-domain learning methods for high-level visual concept classification’. Int. Conf. on Image Processing, 2008.
-
13)
-
8. Yang, J., Yan, R., Hauptmann, A.G.: ‘Cross-domain video concept detection using adaptive SVMs’. Int. Conf. on Multimedia, 2007.
-
14)
-
30. Absil, P.-A., Mahony, R., Sepulchre, R.: ‘Optimization algorithms on matrix manifolds’ (Princeton University Press, 2008).
-
15)
-
35. Torresani, L., Szummer, M., Fitzgibbon, A.: ‘Efficient object category recognition using classemes’. European Conf. on Computer Vision, 2010, pp. 776–789.
-
16)
-
33. Löfberg, J.: ‘YALMIP: a toolbox for modeling and optimization in MATLAB’. Proc. of the CACSD Conf., Taiwan Taipei, 2004.
-
17)
-
31. Wen, Z., Yin, W.: ‘A feasible method for optimization with orthogonality constraints.’, Math. Program., 2013, 142, (1–2), pp. 397–434.
-
18)
-
2. Lichman, M.: , Irvine, CA: University of California, School of Information and Computer Science2013.
-
19)
-
12. Torresani, L., Bergamo, A.: ‘Exploiting weakly-labeled web images to improve object classification: a domain adaptation approach’. Neural Information Processing Systems (NIPS), 2010, pp. 181–189.
-
20)
-
6. Beijbom, O.: ‘Domain adaptation for computer vision applications’, , University of California, San Diego, 2012.
-
21)
-
11. Zen, G., Sangineto, E., Ricci, E., et al: ‘Unsupervised domain adaptation for personalized facial emotion recognition’. Int. Conf. on Multimodal Interaction, 2014, pp. 128–135.
-
22)
-
14. Ma, A.J., Yuen, P.C., Li, J.: ‘Domain transfer support vector ranking for person re-identification without target camera label information’. IEEE Int. Conf. on Computer Vision, 2013, pp. 3567–3574.
-
23)
-
23. Samanta, S., Das, S.: ‘Domain adaptation based on eigen-analysis and clustering, for object categorization’. Int. Conf. on Computer Analysis of Images and Patterns., 2013, LNCS.
-
24)
-
18. Kulis, B., Saenko, K., Darrell, T.: ‘What you saw is not what you get: domain adaptation using asymmetric kernel transforms’. IEEE Conf. on Computer Vision and Pattern Recognition, 2011.
-
25)
-
13. Bruzzone, L., Marconcini, M.: ‘Domain adaptation problems: a dasvm classification technique and a circular validation strategy’, IEEE Trans. Pattern Anal. Mach. Intell., 2010, 32, (5), pp. 770–787.
-
26)
-
27. Pezeshki, A., Scharf, L.L., Chong, E.K.P.: ‘The geometry of linearly and quadratically constrained optimization problems for signal processing and communications’, J. Franklin Inst., 2010, 347, (5), pp. 818–835.
-
27)
-
20. Chattopadhyay, R., Chatapuram Krishnan, N., Panchanathan, S.: ‘Topology preserving domain adaptation for addressing subject based variability in semg signal’. AAAI Spring Symp.: Computational Physiology, 2011.
-
28)
-
4. Shi, X., Fan, W., Ren, J.: ‘Actively transfer domain knowledge’. European Conf. on Machine Learning, 2008.
-
29)
-
34. Bay, H., Ess, A., Tuytelaars, T., et al: ‘Speeded-up robust features (SURF)’, Comput. Vis. Image Underst., 2008, 110, (3), pp. 346–359.
-
30)
-
19. Gopalan, R., Li, R., Chellappa, R.: ‘Unsupervised adaptation across domain shifts by generating intermediate data representations’, IEEE Trans. Pattern Anal. Mach. Intell., 2014, 36, (11), pp. 2288–2302.
-
31)
-
5. Pan, S.J., Yang, Q.: ‘A survey on transfer learning’, IEEE Trans. Knowl. Data Eng., 2010, 22, pp. 1345–1359.
-
32)
-
25. Fernando, B., Habrard, A., Sebban, M., et al: ‘Unsupervised visual domain adaptation using subspace alignment’. Int. Conf. in Computer Vision, 2013.
-
33)
-
10. Yeh, Y., Huang, C., Wang, Y.-C.F.: ‘Heterogeneous domain adaptation and classification by exploiting the correlation subspace’, IEEE Trans. Image Process., 2014, 23, (5), pp. 2009–2018.
-
34)
-
15. Tao, J., Chung, F.-L., Wang, S.: ‘On minimum distribution discrepancy support vector machine for domain adaptation’, Pattern Recognit., 2012, 45, (11), pp. 3962–3984.
-
35)
-
1. Pan, S.J., Tsang, I.W., Kwok, J.T., et al: ‘Domain adaptation via transfer component analysis’, IEEE Trans. Neural Netw., 2011, 22, (2), pp. 199–210.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2015.0322
Related content
content/journals/10.1049/iet-cvi.2015.0322
pub_keyword,iet_inspecKeyword,pub_concept
6
6