Linear Kernel Function

machine-learning

Definition

Linear Kernel Function

The simplest version of a kernel is the linear kernel. If you have data that is already linearly separable and want to use kernel methods for big data sets, a linear kernel has the advantage of requiring less runtime than more complex kernels such as the RBF kernel. The linear kernel is defined as:

$$k(x_i,x_j)=x_i^T{x}_j$$