Abstract
Learning problems for Neural Networks (NNs) has widely been explored from last two decades. Population based algorithms become more focus by researchers because of its nature behavior processing with optimal solution. The population-based algorithms are Ant Colony Optimization (ACO), Artificial Bee Colony (ABC), and recently Hybrid Ant Bee Colony (HABC) Algorithms produced easy way for training NNs. These social based techniques mostly used for finding optimal weight values and over trapping local minima in NNs learning. Typically, NNs trained by a traditional and recognized algorithm called Backpropagation (BP) has difficulties such as trapping in local minima, slow convergence or might fail sometimes. In this research, the new method named Global Hybrid Ant Bee Colony (GHABC) algorithm used to train NNs to recover the BP gaps. The simulation result of a hybrid algorithm evaluates with ordinary ABC, Levenberg-Marquardt (LM) training algorithms. From the investigated results, the proposed GHABC algorithm did get better the learning efficiency for NNs using Boolean Function classification task.