Course Archives Statistical Quality Control & Operations Research Unit
Course:Support Vector Machines
Level: Postgraduate
Time: Currently not offered
Past Exams

Syllabus: Learning machines: attraction and drawbacks of learning, generalization.
SVM--History and Overview.
Loss functions and their risks
SVMs for classification: Hard margin SVMs--- linear SV classifiers for linearly separable data (maximal margin classifiers), Soft margin SVMs—Kernels and reproducing Kernel Hilbert spaces, construction of nonlinear SV classifiers, Linear and Non-linear SVCs in Matlab in Matlab, R (20 hours)
Training of SVMs
Infinite-sample versions of SVMs
Statistical analysis of SVMs
SVM for Regression-Linear and nonlinear SVRs
Live cases with application of SVMs

Reference Texts:
1. Statistical learning Theory: By V.N.Vapnik, John Wiley, New York, 1998.
2. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods: By Nello Cristianini and John Shawe-Taylor, Cambridge University Press, 2000.
3. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond: By B. Sch¨olkopf and A. J. Smola, MIT Press, Cambridge, MA, 2002.
4. Support Vector Machines: Theory and Applications: By Lipo Wang (Ed.), Springer- Verlag, New York, 2005.
5. Support Vector Machines for Pattern Classification: By Shigeo Abe, Springer-Verlag, London, 2005.

Top of the page

Past Exams
Supplementary and Back Paper

Top of the page

[Indian Statistical Institute]