Which of the following are true statements about nearest neighbors classifiers? (Select all that apply.)

a) Nearest neighbors classifiers work best in applications in which the decision boundaries are highly regular and can be well approximated by intersecting hyperplanes.

b) Increasing the number of neighbors, k, increases the smoothness of the decision boundary found by a nearest neighbors classifier.

c) Since nearest neighbors classifiers require no model to be fit, the computational load is relatively light compared to model-based methods.