Learning the Kernels for Support Vector Machines

Speaker: Shai Ben-David

Support Vector machines (SVM's) is one of the most useful and widely applicable machine learning techniques. Each concrete application of SVMs depends on a successful choice of a "kernel matrix". So far, most of the work in this area has focused on developing a variety of kernels and efficient algorithms for employing SVMs with these kernels. Relatively little research attention has been given to the question of how to pick a suitable kernel for any particular learning task at hand.

In this work, we analyze exactly that issue. We show that to a certain extent the problem of choosing a kernel can be dealt with as another learning task, for a moderate cost in terms of the number of needed training examples.

Rather than aiming at the SVM specialists among you, I wish to use the opportunity to provide a general introduction to the useful ideas of support vector machines and kernel-based learning. I shall therefore assume no prior background on these issues, and will not go into the technical details of the new results.

The new results about learning the kernel are joint work with Nati Srebro from U of T.