On Comparative Study of Neural Networks' Application for Three Behavioral Learning Systems' Activities Versus Swarm Intelligence Performance.

Abstract

Artificial neural network modeling; Swarm Intelligence; ant colony system; traveling salesman problem; computational biology. This piece of research addresses an interesting comparative analytical study. Which considers two concepts of diverse algorithmic computational intelligence approaches related tightly with Neural and Non-Neural Systems. The first algorithmic intelligent approach concerned with observed obtained practical results after three neural animal systems' activities. Namely, they are Pavlov's, and Thorndike's experimental work. Besides a mouse's trial during its movement inside figure of eight (8) maze, to reach an optimal solution for reconstruction problem. Conversely, second considers algorithmic Swarm Intelligent (SI) approach originated from resulting activities of Non-Neural Ant Colony System (ACS). These results obtained after reaching an optimal solution while solving Traveling Sales-man Problem (TSP). Interestingly, the effect of the increase for agents' number (either neurons or ants) on learning systems' performance shown to be in agreement of each other for both (neural and non-neural) systems. Finally, performance of both intelligent learning paradigms shown to be in agreement with learning convergence process searching for Least Mean Square LMS error algorithm. That's during application of training phase of three Artificial Neural Network (ANN) models. Accordingly, adopted ANN modeling is a relevant and realistic tool to investigate observations and analyze performance for both selected computational intelligence (biological behavioral learning) systems.