SVM: Classical and Quantum Kernel Classifiers
This SVM series builds from first principles to a head-to-head comparison of classical and quantum kernel classifiers. Each notebook follows the same pattern: mathematics first, then code.
1. Support Vector Machines
Derive the maximum-margin hyperplane, read the primal and dual forms of the SVM, and control the soft-margin trade-off with \(C\).
Prerequisites: vectors, inner products, and basic Python.
2. Kernels in Support Vector Machines
Understand feature maps, the kernel trick, and the kernel SVM decision function through polynomial and RBF examples.
Prerequisites: 1. Support Vector Machines.
3. SVM: Classical and Quantum Kernel Classifiers
Compare a classical RBF kernel and a quantum feature-map kernel side by side on the same dataset using the same support-vector learning rule.
Prerequisites: 2. Kernels in SVMs.
Refreshers
1.1 SVM Margin Geometry in Two Features
Draw the decision and margin lines and prove that the full geometric margin is \(2/\lVert\mathbf{w}\rVert\).
Companion to: 1. Support Vector Machines, Section 2.