The Kernel Trick: Nonlinear Boundaries for Free

Classical models built from the math you already know

The support vector machine from Margins & the Support Vector Machine only ever draws a straight line (or, in more than two dimensions, a flat hyperplane). That is a real limitation. Plenty of data refuses to sit in two neat, straight-line-separable piles.

🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.

▶ The Kernel Trick: Nonlinear Boundaries for Free
← Margins & the Support Vector MachineMulticlass, the Confusion Matrix & Metrics →