Efficient learning algorithms for classification of data by a nonlinear decision boundary

In this thesis, we study the classification of data by a nonlinear classification boundary in the case the decision boundary is determined by the inverse of a given function. We propose and investigate efficient machine learning algorithms based on the Gabriel Edited Set, which is an approximate decision-boundary consistent reduced set. The algorithms attempts to account for the geometry of the equivalence classes determined by the inverse of the function. We numerically investigate the behaviour and accuracy of the algorithms on multiple examples.