Recent Publications and Preprints

Full list of publications on both Statistical Machine Learning and Mathematics (Harmonic Analysis), see my CV

• Y. Ying and D. X. Zhou. Online pairwise learning algorithms. Preprint, 2014.

• Q. Cao, Y. Ying, M. Pontil, and Peng Li. Similarity metric learning for unconstrained face verification and person re-identification. Submitted for publication, 2014.

• Y. Lei and Y. Ying. Generalization analysis for multi-modal metric learning. Submitted for publication, 2014.

• M. Rogers, C. Campbell and Y. Ying. Probabilistic inference of biological networks via data integration. To appear in BioMed Research International, 2014.

• J. Bohne, Y. Ying, S. Gentric and M. Pontil, Large margin local metric learning . European Conference on Computer Vision (ECCV), 2014.

• Z. C. Guo and Y. Ying, Guaranteed classification via regularised similarity learning. Neural Computation , Vol. 26(3), 2014.

• Q. Cao, Y. Ying and Peng Li, Similarity metric learning for face recognition. IEEE International Conference on Computer Vision (ICCV), 2013.

• Q. Cao, Z. C. Guo and Y. Ying, Generalization bounds for metric and similarity learning. arXiv preprint, version 2 (under revision for Machine Learning Journal) , 2013.

• Y. Ying, Multi-task coordinate gradient learning. ICML workshop on "object, functional and structured data: towards next generation kernel-based methods", 2012.

• Q. Cao, Y. Ying and P. Li, Distance metric learning revisited. ECML-PKDD , 2012. (version 1, April 2012)

• Y. Ying and P. Li, Distance metric learning with eigenvalue optimization. Journal of Machine Learning Research (Special topics on kernel and metric learning) , 13:1-26, 2012.

See also the application of our method on face verification data (LFW) .

• C. Campbell and Y. Ying, Learning with Support Vector Machines. Morgan & Claypool Publishers, 2011.

• Y. Ying and K. Huang and C. Campbell, Distance metric learning with sparse regularization. Technical Report, University of Exeter, September, 2010.

• Y. Ying, Q. Wu and C. Campbell, Learning the coordinate gradients, to appear in Advances in Computational Mathematics, 2010.

• Y. Ying and C. Campbell, Rademacher chaos complexity for learning the kernel, Neural Computation, Vol. 22(11), 2010. (version 1, October 2008)

This second reversion is a substantial extension of the COLT (2009) conference paper: Generalization bounds for learning the kernel. In particular, we provided a self-contained proof for bounding the Rademacher chaos complexity by metric entropy integrals and also corrected the inaccurate claim on generalization bounds derived from the covering number approach (appeared at the end of Section 3 of the COLT conference version).

• K. Huang, Y. Ying and C. Campbell, Generalized sparse metric learning with relative comparisons, Journal of Knowledge and Information Systems (KAIS), 2010.

• K. Huang, Y. Ying and C. Campbell, GSML: A unified framework for sparse metric learning, IEEE International Conference on Data Mining (ICDM), 2009.

• Y. Ying, K. Huang and C. Campbell, Sparse metric learning via smooth optimization, Advances in Neural Information Processing Systems (NIPS), 2009.

• Y. Ying, C. Campbell and M. Girolami, Analysis of SVM with indefinite kernels, Advances in Neural Information Processing Systems (NIPS), 2009. spotlight slide

MATLAB code (version 1)

• Y. Ying, K. Huang and C. Campbell, Enhanced protein fold recognition through a novel data integration approach, BMC Bioinformatics (Open access), (2009) 10:267.

See its short paper Information theoretic kernel integration , NIPS workshop on Learning from multiple sources, 2009. Presentation

• Y. Ying and C. Campbell, Generalization bounds for learning the kernel, Proceedings of the 22nd Annual Conference on Learning Theory (COLT), 2009.

• Y. Ying, C. Campbell, T. Damoulas, and M. Girolami, Class prediction from disparate biological data sources using a simple multi-class multi-kernel algorithm, In 4th IAPR International Conference on Pattern Recognition in Bioinformatics, 2009. (Was preprint, 2008)

• T. Damoulas, Y. Ying, M. Girolami, and C. Campbell, Inferring sparse kernel combination and relevance vectors: an application to subcelluar localization of proteins , International Conference on Machine Learning and Applications (ICMLA), 2008.

• Y. Ying and C. Campbell, Learning coordinate gradients with multi-task kernels , Proceedings of the 21st Annual Conference on Learning Theory (COLT), 2008.

MATLAB code available under request.

• P. Agius, Y. Ying and C. Campbell, Bayesian unsupervised learning with multiple data types , Statistical Applications in Genetics and Molecular Biology, Vol. 8: Iss. 1, 2009. (Was Technical Report, January, 2008)

• P. Li, Y.Ying and C. Campbell, A variational approach to semi-supervised clustering, Proceedings of ESSAN , 2009. (Was preprint, 2008)

• Y. Ying, P. Li and C. Campbell, A marginalized variational Bayesian approach to the analysis of array data, BMC Proceedings of the Workshop on Machine Learning in Systems Biology, 2007.

• A. Argyriou, C. A. Micchelli, M. Pontil, and Y. Ying, A spectral regularization framework for multi-task structure learning, Advances in Neural Information Processing Systems (NIPS), 2007.

• A. Caponnetto, C. A. Micchelli, M. Pontil, and Y. Ying, Universal multi-task kernels, Journal of Machine Learning Research, 9 (2008), 1615-1646. (Was Technical Report, University College London, December 2006)

Characterization of universal matrixed-valued (and more general operator-valued) kernels. These characterizations are highlighted with numerous examples of paractical importance in multi-task learning.

• M. Pontil and Y. Ying, Online gradient descent learning algorithms, Foundations of Computational Mathematics, 5 (2008), 561-596. (Was Technical Report, University College London, 2005)

Generalization analysis of Online Stochastic Gradient Descent algorithms in reproducing kernel Hilbert space. In particular, we show that their error rates are competitive with offline regularization algorithms.

• Y. Ying and D.X. Zhou, Online regularized classification algorithms, IEEE Trans. Inform. Theory (regular paper), 11 (2006), 4775-4788.

• Y. Ying and D.X. Zhou, Learnability of Gaussians with flexible variances , Journal of Machine Learning Research, 8 (2007), 249-276. (Was Technical Report, City University of Hong Kong, 2004)

Characterization of statistical consistency for learning the kernel algorithms by the V-gamma dimension of the set of candidate kernels.

• Q. Wu, Y. Ying and D.X. Zhou, Multi-kernel regularized classifiers, Journal of Complexity, 2006. (Was preprint, 2004)

• Y. Ying, Convergence analysis of online algorithms, Advances in Computational Mathematics, 27 (2007), 273-291. (Was preprint, City University of Hong Kong, 2005).

• Q. Wu, Y. Ying and D.X. Zhou, Learning rates of least-square regularized regression, Foundations of Computational Mathematics, 6 (2006), 171-192.

• Y. Ying, McDiarmid's inequalities of Bernstein and Bennett forms, Technical Report, City University of Hong Kong, 2004.

• Q. Wu, Y. Ying and D.X. Zhou, Learning theory: from regression to classification, Topics in Multivariate Approximation and Interpolation, K.Jetter et.al., Editors, (2004), 101--134.

• D.R. Chen, Q. Wu, Y. Ying and D.X. Zhou, Support vector machine soft margin classifiers: error analysis, Journal of Machine Learning Research, 5 (2004).