Solving a classification task using Functional Link Neural Networks with modified Artificial Bee Colony

Abstract

Classification is one of the most frequent studies in the area of Artificial Neural Network (ANNs). The ANNs are capable of generating complex mapping between the input and the output space and thus these networks can form arbitrarily complex nonlinear decision boundaries. One of the best known types of ANNs is the Multilayer Perceptron (MLP). However, MLP usually requires a large amount of available measures in order to achieve good classification accuracy. To overcome this, a Functional Link Neural Networks (FLNN) which has single layer of trainable connection weights is used. The standard method for tuning the weight in FLNN is using a Backpropagation (BP-learning) algorithm. Still, BP-learning algorithm has difficulties such as trapping in local optima and slow convergence that can affect the FLNN performance. In this work, an Artificial Bee Colony (ABC) algorithm known to have good exploration and exploitation capabilities in searching optimal weight is used to recover the BP-learning drawbacks. With modifications on the employed and onlooker bee's foraging behavior, the implementation of the modified ABC as a learning scheme for FLNN has resulted in better accuracy rate for solving classification tasks.