Evolutionary combinatorial optimization for recursive supervised learning with clustering

The idea of using a team of weak learners to learn a dataset is a successful one in literature. In this paper, we explore a recursive incremental approach to ensemble learning. In this paper, patterns are clustered according to the output space of the problem, i.e., natural clusters are formed based on patterns belonging to each class. A combinatorial optimization problem is therefore formed, which is solved using evolutionary algorithms. The evolutionary algorithms identify the "easy" and the "difficult" clusters in the system. The removal of the easy patterns then gives way to the focused learning of the more complicated patterns. Incrementally, neural networks are added to the ensemble to focus on solving successively difficult examples. The problem therefore becomes recursively simpler. Over fitting is overcome by using a set of validation patterns along with a pattern distributor. An algorithm is also proposed to use the pattern distributor to determine the optimal number of recursions and hence the optimal number of weak learners for the problem. In this paper, we show that the generalization accuracy of the proposed algorithm is always better than that of the underlying weak learner. Empirical studies show generally good performance when compared to other state-of-the-art methods.