A Kernel Function is a function that implements a notion of similarity between points in a given space, by implicitly calculating the dot product between the mappings of those points in a higher dimensional space. ML methods that use kernel functions are referred to as kernel methods, or kernel machines -- the Support Vector Machine (SVM) with a Radial Basis Function (RBF) kernel is one of the most popular. Kernel functions are often used to empower linear models to solve non-linear problems -- this is known as the "kernel trick", and SVMs with RBF kernels are one such case.