Abstract
Image annotation is a typical area where there are multiple types of attributes associated with each individual image. In order to achieve better performance, it is important to develop effective modeling by utilizing prior knowledge. In this article, we extend the graph regularization approaches to a more general case where the regularization is imposed on the factorized variational distributions, instead of posterior distributions implicitly involved in EM-like algorithms. In this way, the problem modeling can be more flexible, and we can choose any factor in the problem domain to impose graph regularization wherever there are similarity constraints among the instances. We formulate the problem formally and show its geometrical background in manifold learning. We also design two practically effective algorithms and analyze their properties such as the convergence. Finally, we apply our approach to image annotation and show the performance improvement of our algorithm.
- Attias, H. 2000. A variational Bayesian framework for graphical models. Adv. Neural Info. Proc. Syst. 12, 209--215.Google Scholar
- Barnard, K., Duygulu, P., Forsyth, D., de Freitas, N., Blei, D. M., and Jordan, M. I. 2003. Matching words and pictures. J. Mach. Learn. Res. 3, 1107--1135. Google ScholarDigital Library
- Belkin, M. and Niyogi, P. 2001. Laplacian eigenmaps and spectral techniques for embedding and clustering. In Proceedings of the Conference on Advances in Neural Information Processing Systems 14, 585--591.Google Scholar
- Belkin, M., Niyogi, P., and Sindhwani, V. 2006. Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. J. Mach. Learn. Res. 7, 2399--2434. Google ScholarDigital Library
- Bilmes, J. 2004. On virtual evidence and soft evidence in Bayesian networks. Tech. rep. UWEETR-2004-0016, Department of EE, University of Washington.Google Scholar
- Bishop, C. M. 2007. Pattern Recognition and Machine Learning. Springer.Google Scholar
- Blei, D., Ng, A., and Jordan, M. 2003. Latent dirichlet allocation. J. Mach. Learn. Res. Google ScholarDigital Library
- Blei, D. M. and Jordan, M. I. 2003. Modeling annotated data. In Proceedings of the ACM International Conference on Research and Development in Informaion Retrieval (ACM SIGIR). 127--134. Google ScholarDigital Library
- Bousquet, O., Boucheron, S., and Lugosi, G. 2003. Introduction to statistical learning theory. In Advanced Lectures on Machine Learning. 169--207.Google Scholar
- Cai, D., Mei, Q., Han, J., and Zhai, C. 2008. Modeling hidden topics on document manifold. In Proceedings of the ACM Conference on Information and Knowledge Management (CIKM'08). 911--920. Google ScholarDigital Library
- Chang, E. and Sychay, G. 2003. CBSA: Content-based soft annotation for multimodal image retrieval using bayes point machines. IEEE Trans. Circ. Syst. Video Tech. 13, 26--38. Google ScholarDigital Library
- Chang, J. and Blei, D. 2009. Relational topic models for document networks. In Proceedings of Conference on AI and Statistics.Google Scholar
- Chung, F. R. K. 1997. Spectral Graph Theory. Regional Conference Series in Mathematics, vol. 92. AMS.Google Scholar
- Csurka, G., Dance, C. R., Fan, L., Willamowski, J., and Bray, C. 2004. Visual categorization with bags of keypoints. In Proceedings of the Workshop on Statistical Learning in Computer Vision (ECCV). 1--22.Google Scholar
- Hastie, T., Tibshirani, R., and Friedman, J. H. 2001. The Elements of Statistical Learning. Springer-Verlag.Google Scholar
- He, X. 2010. Laplacian regularized d-optimal design for active learning and its application to image retrieval. Trans. Img. Proc. 19, 1, 254--263. Google ScholarDigital Library
- He, X., Cai, D., Shao, Y., Bao, H., and Han, J. 2009a. Laplacian regularized gaussian mixture model for data clustering. IEEE Trans. Knowl. Data Engin. Google ScholarDigital Library
- He, X., Ji, M., and Bao, H. 2009b. Graph embedding with constraints. In Proceedings of the 21st International Jont Conference on Artifical Intelligence (IJCAI'09). Morgan Kaufmann Publishers Inc., San Francisco, CA, 1065--1070. Google ScholarDigital Library
- He, X., Ji, M., and Bao, H. 2009c. A unified active and semi-supervised learning framework for image compression. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 65--72.Google Scholar
- He, X., Ji, M., and Bao, H. 2009d. A unified active and semi-supervised learning framework for image compression. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 65--72.Google Scholar
- Hofmann, T. 1999. Probabilistic latent semantic indexing. In Proceedings of the ACM International Conference on Research and Development in Information Retrieval (SIGIR'05). ACM, 50--57. Google ScholarDigital Library
- Hofmann, T. 2001. Unsupervised learning by probabilistic latent semantic analysis. Mach. Learn. 42, 1-2, 177--196. Google ScholarDigital Library
- Jordan, M. I., Ed. 1999. Learning in Graphical Models. MIT Press, Cambridge, MA. Google ScholarDigital Library
- Jordan, M. I., Ghahramani, Z., Jaakkola, T. S., and Saul, L. K. 1999. An introduction to variational methods for graphical models. In Learning in Graphical Models. MIT Press, Cambridge, MA, 105--161. Google ScholarDigital Library
- Li, L.-J., Socher, R., and Fei-Fei, L. 2009. Towards total scene understanding:classification, annotation and segmentation in an automatic framework. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Google Scholar
- Mei, Q., Cai, D., Zhang, D., and Zhai, C. 2008. Topic modeling with network regularization. In Proceedings of the ACM International Conference on World Wide Web (WWW'08). 101--110. Google ScholarDigital Library
- Minka, T. P. 2003. Estimating a dirichlet distribution. http://research.microsoft.com/minka.Google Scholar
- Monay, F. and Gatica-Perez, D. 2003. On image auto-annotation with latent space models. In Proceedings of the ACM International Conference on Multimedia (SIGMM'03). 275--278. Google ScholarDigital Library
- Monay, F. and Gatica-Perez, D. 2004. pLSA-based image auto-annotation: constraining the latent space. In Proceedings of the ACM International Conference on Multimedia (SIGMM'04). 348--351. Google ScholarDigital Library
- Neal, R. M. and Hinton, G. E. 1999. A view of the em algorithm that justifies incremental, sparse, and other variants. In Learning in Graphical Models. 355--368. Google ScholarDigital Library
- Nocedal, J. and Wright, S. 2006. Numerical Optimization. Springer.Google Scholar
- Pearl, J. 1988. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann Publishers Inc., San Francisco, CA. Google ScholarDigital Library
- Rubner, Y., Tomasi, C., and Guibas, L. J. 2000. The earth mover's distance as a metric for image retrieval. Int. J. Comput. Vis. 40, 2, 99--121. Google ScholarDigital Library
- Shao, Y., Zhou, Y., He, X., Cai, D., and Bao, H. 2009. Semi-supervised topic modeling for image annotation. In Proceedings of the 17th ACM International Conference on Multimedia (MM'09). ACM, New York, 521--524. Google ScholarDigital Library
- Si, S., Tao, D., and Geng, B. 2010. Bregman divergence-based regularization for transfer subspace learning. IEEE Trans. Knowl. Data Engin. 22, 929--942. Google ScholarDigital Library
- Song, D. and Tao, D. 2010. Biologically inspired feature manifold for scene classification. Trans. Img. Proc. 19, 1, 174--184. Google ScholarDigital Library
- Stephen, E. E., Fienberg, S., and Lafferty, J. 2004. Mixed membership models of scientific publications. In Proc. National Acad. Sci.Google Scholar
- Tao, D., Li, X., Wu, X., and Maybank, S. J. 2009. Geometric mean for subspace selection. IEEE Trans. Patt. Anal. Mach. Intell. 31, 2, 260--274. Google ScholarDigital Library
- Vapnik, V. N. 1995. The Nature of Statistical Learning Theory. Springer-Verlag, Berlin. Google ScholarDigital Library
- Vapnik, V. N. 1998. Statistical Learning Theory. Wiley.Google Scholar
- Winn, J. and Bishop, C. M. 2005. Variational message passing. J. Mach. Learn. Res. 6, 661--694. Google ScholarDigital Library
- Xing, E. P., Jordan, M. I., and Russell, S. J. 2003. A generalized mean field algorithm for variational inference in exponential families. In Proceedings of the International Conference on Uncertainty in Artificial Intelligence. 583--591. Google ScholarDigital Library
- Zhang, R., Zhang, Z. M., Li, M., Ma, W.-Y., and Zhang, H.-J. 2005. A probabilistic semantic model for image annotation and multi-modal image retrieva. In Proceedings of the IEEE International Conference on Computer Vision (ICCV'05). 846--851. Google ScholarDigital Library
- Zhou, T., Tao, D., and Wu, X. 2010. Manifold elastic net: A unified framework for sparse dimension reduction. Data Min. Knowl. Disc. Google ScholarDigital Library
- Zhu, X., Lafferty, J. and Ghahramani, Z. 2005. Semi-supervised learning using gaussian fields and harmonic functions. In Proceedings of the International Conference on Machine Learning (ICML'05).Google Scholar
Index Terms
- Variational inference with graph regularization for image annotation
Recommendations
Semi-supervised topic modeling for image annotation
MM '09: Proceedings of the 17th ACM international conference on MultimediaWe propose a novel technique for semi-supervised image annotation which introduces a harmonic regularizer based on the graph Laplacian of the data into the probabilistic semantic model for learning latent topics of the images. By using a probabilistic ...
Stochastic variational inference
We develop stochastic variational inference, a scalable algorithm for approximating posterior distributions. We develop this technique for a large class of probabilistic models and we demonstrate it with two probabilistic topic models, latent Dirichlet ...
L1-norm Laplacian support vector machine for data reduction in semi-supervised learning
AbstractAs a semi-supervised learning method, Laplacian support vector machine (LapSVM) is popular. Unfortunately, the model generated by LapSVM has a poor sparsity. A sparse decision model has always been fascinating because it could implement data ...
Comments