What is the 'Kernel' in SVM?

What is the 'Kernel' in SVM?

Approach: Define the function that calculates similarity in higher dimensions.

A function that computes the dot product of two vectors in some (possibly very high-dimensional) feature space. It allows SVMs to solve non-linear problems without explicitly transforming data.

Follow-up: What is the RBF kernel?

Key concepts: dot product, mapping, non-linearity.