Geometry of margins and feature spaces
Linear classifiers and margins
Support vectors and margin maximization
The primal and dual SVM formulation
The kernel trick: implicit feature maps
Common kernels: RBF, polynomial, string
SVMs vs. neural networks today