A linear learner has high bias, because when the frontier between two classes is not a hyperplane the learner is unable to induce it Can someone explain to me what this sentence means:

A linear learner has high bias, because when the frontier between two classes is not a hyperplane the learner is unable to induce it.

The quote is from here in the 5th section
I understand what bias and hyperplane means so I guess I'm confused about the terms linear learner and frontier. I've search on the internet but I still confused as to what these terms mean
 A: There the author is talking about binary classification. We have a collection of training data, some labelled as class $A$ and the rest labelled as class $B$. The goal is to come up with some kind of decision function for assigning class labels to new unlabelled data points that we might see in the future.
One way one might go about this is to fit some kind of curve such that all the points labelled $A$ are on one side of the curve, and all the points labelled $B$ are on the other side. Then when you get a new unlabelled input point, you just check which side of the surface it falls on and guess a class label accordingly.
For simplicity, we might restrict the set of curves we consider to the set of hyperplanes. Then the goal becomes finding a hyperplane which separates the $A$'s from the $B$'s. This is what is meant by a linear learner in this case, since hyperplanes are linear. The  frontier is the true curve which separates the $A$'s from the $B$'s (which we have only incomplete information about via the training data). If the shape of the frontier is not a hyperplane, then even the hyperplane which separates the data the best will not separate it very well. We would say there is bias, because the hyperplane that the algorithm selects will inevitably be "far from" the true separating curve.
To put it another way, a linear learner is just a learning algorithm for binary classification whose hypothesis space (i.e. set of curves considered) is the set of hyperplanes. The frontier is the true, unknown (to us) curve which separates the two classes.
