Distance Metric Learning via Sparse Regularization, Yiming Ying, Kaizhu Huang and Colin Campbell Journal submission.

Efficient SVM Classification with Indefinite Kernels, Yiming Ying and Colin Campbell. Journal submission.

Rademacher chaos complexity for learning the kernel, Yiming Ying and Colin Campbell. Neural Computation, 22, 2010, p. 2858-2886.

Download the pdf.

An earlier paper on this approach (Bounds for Learning the Kernel: Rademacher Chaos Complexity) was released in 2008 and you can download the original pdf. This latest version is an extension of our COLT conference paper: Generalization bounds for learning the kernel. We now provide a self-contained proof for bounding the Rademacher chaos complexity by metric entropy integrals and also correct the claim about generalization bounds derived from the covering number approach (which appeared at the end of Section 3 of the COLT conference version).

Generalized sparse metric learning with relative comparisons. Kaizhu Huang, Yiming Ying and Colin Campbell. Knowledge and Information Systems (KAIS), volume 28, issue 1, 2011, pages 25-45.

Download the pdf

Sparse Metric Learning via Smooth Optimization. Yiming Ying, Kaizhu Huang and Colin Campbell. Advances in Neural Information Processing Systems 22, 2009, p. 2214-2222.

Download the pdf

Analysis of SVM with Indefinite Kernels. Yiming Ying, Colin Campbell and Mark Girolami. Advances in Neural Information Processing Systems 22, 2009, p. 2205-2213.

Download the pdf

GSML: A Unified Framework for Sparse Metric Learning. Kaizhu Huang, Yiming Ying and Colin Campbell. Proceedings IEEE International Conference on Data Mining, ICDM 2009.

Download the pdf

Generalization Bounds for Learning the Kernel. Yiming Ying and Colin Campbell. Proceedings: Computational Learning Theory (COLT), 2009.

Download the pdf

Learning the Coordinate Gradients. Yiming Ying, Qiang Wu and Colin Campbell. Advances in Computational Mathematics, 2010.

Download the pdf

Learning coordinate gradients with multi-task kernels. Yiming Ying and Colin Campbell. Proceedings: Computational Learning Theory (COLT), 2008, p. 217-228.

View the abstract. Download the pdf

Kaizhu Huang, Zenglin Xu, Irwin King, Michael R. Lyu and Colin Campbell. Supervised Self-taught Learning: Actively Transferring Knowledge from Unlabeled Data, Proceedings of IJCNN09, 2009, p. 1272-1277.

Query Learning with Gaussian Processes. Simon Rogers and Colin Campbell. Proceedings of the 2003 UK Workshop on Computational Intelligence, 2003, pages 45-52.

A Linear Programming Approach to Novelty Detection. Colin Campbell and Kristin P. Bennett. Advances in Neural Information Processing Systems, vol. 13, MIT press, Cambridege, MA, 2001, p. 395-401.

View the abstract or download the pdf or postscript.

An Introduction to Kernel Methods. C. Campbell. Radial Basis Function Networks: Design and Applications. R.J. Howlett and L.C. Jain (eds). Physica Verlag, Berlin, 2000, Ch. 7 p. 155-192.

View the abstract or download the pdf file.

Machine Learning Strategies for Complex Tasks. Colin Campbell, Theodoros Evgeniou, Bernd Heisele and Massimiliano Pontil. Proceedings of First IEEE-RAS International Conference on Humanoid Robots, MIT, 2000, Springer Verlag, 13 pages..

View the abstract or download the pdf or postscript

Query Learning with Large Margin Classifiers. C. Campbell, N. Cristianini and A. Smola. Proceedings of the 17th International Conference on Machine Learing, (ICML2000, Stanford, CA, 2000), Morgan Kaufmann, p. 111-118..

View the abstract or download the pdf or postscript.

Support Vector Machine Classification and Validation of Cancer Tissue Samples using Microarray Expression Data. T. Furey, N. Cristianini, N. Duffy, D. Bednarski, Michel Schummer and D. Haussler Bioinformatics, 2000, 16:906-914

View details

Algorithmic Approaches to Training Support Vector Machnies: A Survey. C. Campbell. Proceedings of ESANN2000 (D-Facto Publications, Belgium, 2000) p. 27-36.

View the abstract or download the pdf or postscript.

Kernel Methods: A Survey of Current Techniques. C. Campbell. Neurocomputing 2002, 48: 63-84.

Download the pdf

Bayes Point Machines. Ralf Herbrich, Thore Graepel, Colin Campbell. Journal of Machine Learning Research 1(2001) p. 245-279.

Download the pdf
or compressed postScript

Robust Bayes Point Machines. R. Herbrich, Th. Graepel and C. Campbell. Proceedings of ESANN2000 (D-Facto Publications, Belgium, 2000) p. 49-54.

View the abstract or download the pdf or postscript.

Large Margin DAGs for Multiclass Classification J. Platt, N. Cristianini and J. Shawe-Taylor. In Advances in Neural Information Processing Systems, Vol. 12 MIT Press, 2000, p. 547-553.

View the abstract or download the pdf or postscript.

Margin Distribution and Soft Margin. J. Shawe-Taylor and N. Cristianini. In: A.J. Smola, P. Bartlett, B. Schoelkopf, and C.Schuurmans (editors), Advances in Large Margin Classifiers, Chapter 2. MIT Press, 1999.

View the abstract or download the pdf or postscript.

Knowledge-based Analysis of Microarray Gene Expression Data using Support Vector Machines. M. Brown, W. Grundy, D. Lin, N. Cristianini C. Sugnet, T. Furey, M. Ares Jr., D. Haussler Proceedings of the National Academy of Sciences 97(1) p. 262-267.

View details

Controlling the Sensitivity of Support Vector Machines. K. Veropoulos, C. Campbell and N. Cristianini. In Proceedings of the International Joint Conference on Artificial Intelligence, Stockholm, Sweden, 1999 (IJCAI99), Workshop ML3, p. 55-60.

View the abstract or download the pdf or postscript.

Bayesian Voting Schemes and Large Margin Classifiers. N. Cristianini and J. Shawe-Taylor. In: B. Schoelkopf, C. Burges and A. Smola (editors), Advances in Kernel Methods - Support Vector Learning, Chapter 5 (p. 55-68). MIT Press, 1999.

View the abstract or download the pdf or postscript.

Margin Distribution Bounds on Generalisation. J. Shawe-Taylor and N. Cristianini. In: Lecture Notes in Artificial Intelligence, 1572, Computational Learning Theory (Proceedings of Eurocolt 1999). p. 263-273.

View the abstract or download the pdf or postscript.

Robust Bounds on Generalisation from the Margin Distribution. J. Shawe-Taylor and N. Cristianini. Submitted to IEEE Transactions on Information Theory..

View the abstract or download the pdf or postscript.

Dynamically Adapting Kernels in Support Vector Machines. N. Cristianini, C. Campbell and J. Shawe-Taylor. In M. Kearns, S. Solla and D. Cohn (editors), Advances in Neural Information Processing Systems, Vol. 11. MIT Press, 1999, p. 204-210.

View the abstract or download the pdf or postscript.

Multiplicative Updatings for Support Vector Machines. N. Cristianini, C. Campbell and J. Shawe-Taylor. In: Proceeding of ESANN 99 (D-Facto Publications, Belgium, 1999) p. 189-194.

View the abstract or download the pdf or postscript.

Enlarging the Margin in Perceptron Decision Trees. K. Bennett, N. Critianini, J. Shawe-Taylor and D. Wu. Machine Learning 41 (3), 2000, pp. 295-313.

View the abstract or download the pdf or postscript.

Large Margin Decision Trees for Induction and Transduction. D. Wu, K. Bennett, N. Cristianini and J. Shawe-Taylor. In: Proceedings of the Sixteenth International Conference on Machine Learning (ICML99).

View the abstract or download the pdf or postscript.

Bayes Point Machines: Estimating the Bayes Point in Kernel Space. R. Herbrich, Th. Graepel and C. Campbell. In: Proceedings of the International Joint Conference on Artificial Intelligence, Stockholm, Sweden, 1999 (IJCAI99). Workshop ML3, p.23-27

View the abstract or download the pdf or postscript.

Bayesian Learning in Reproducing Kernel Hilbert Spaces: The Usefulness of the Bayes Point. R. Herbrich, Th. Graepel and C. Campbell. Technical Report (1999). Longer version of the above paper: comments on this paper are welcome.

View the abstract or download the pdf or postscript.

Further Results on the Margin Distribution. J. Shawe-Taylor and N. Cristianini. Proceedings of COLT'99, Morgan Kaufmann, p. 278-285.

View the abstract or download the pdf or postscript.

Data Dependent Structural Risk Minimization for Perceptron Decision Trees. J. Shawe-Taylor and N. Cristianini. In: M. Jordan, M. Kearns and S. Solla, Advances in Neural Information Processing Systems, Vol. 10 (p. 336-342). MIT Press, 1998.

View the abstract or download the pdf or postscript.

Bayesian Classifiers are Large Margin Hyperplanes in a Hilbert Space. N. Cristainini, J. Shawe-Taylor and P. Sybacek. In: J. Shavlik (editor), Proceedings of the Fifteenth International Conference on Machine Learning, (1998) p. 109-117.

View the abstract or download the pdf or postscript.

Bayesian Classifiers are Large Margin Hyperplanes in a Hilbert Space. N. Cristianini, J. Shawe-Taylor and P. Sykacek. Journal submission (1998).

View the abstract or download the pdf or postscript.

Simple Training Algorithms for Support Vector Machines. C. Campbell and N. Cristianini. Technical Report: describes KA algorithm, including how to implement a bias (same solution as SVM via QP).

Download the pdf or postscript

The Kernel-Adatron: a Fast and Simple Learning Procedure for Support Vector Machines. Th. Frieß, N. Cristianini and C. Campbell. (First paper on KA algorithm but contains a number of typographical errors). In: J. Shavlik (editor), Proceedings of the Fifteenth International Conference on Machine Learning (1998) p. 188-196.

Large Margin Classification Using the Kernel Adatron Algorithm. C. Campbell, Th. Frieß and N. Cristianini. In: L. Xu, L.W. Chan and A. Fu (editors), IDEAL'98: Intelligent Data Engineering and Learning, Springer, 1998 p. 355-362.