Hypothesis
$$ h_\theta(x) = \begin{cases} 1 & \text{if} \ \theta^Tx \geq 0 \\ 0 & \text{otherwise} \end{cases} $$
Cost Function
$$ J(\theta) = C\sum_{i=1}^m y^{(i)} \ \text{cost}_1(\theta^Tx^{(i)}) + (1 - y^{(i)}) \ \text{cost}_0(\theta^Tx^{(i)}) + \dfrac{1}{2}\sum_{j=1}^n \theta^2_j $$ where $$ C = \text{regularization parameter} = \frac{1}{\lambda} $$
Algorithms
Use libsvm and choose parameters Larger \(C\) gives higher variance Other (non-linear) kernels have extra parameters Gaussian: larger \(\sigma^2\) gives higher bias