Final answer:
In support vector machines, a fat separator refers to the margin, and there are infinitely many choices for it, which helps to regularize the capacity of the model by preventing overfitting.
Step-by-step explanation:
In the context of Support Vector Machines (SVM), if you have to place a fat separator (margin) between classes, you have infinitely many choices because you can select the hyperplane that maximizes the margin between the classes in an infinite-dimensional space.
Therefore, the capacity of the model has been regularized. The regularization is due to the fact that among the numerous possible separators, the SVM algorithm selects the one with the largest margin, essentially simplifying the decision boundary which, in turn, helps to prevent overfitting.