Picking up after last week's episode about maximal margin classifiers, this week we'll go into the kernel trick and how that (combined with maximal margin algorithms) gives us the much-vaunted support vector machine.