Comparative Performance Analysis and Evaluation for One Selected Behavioral Learning System versus an Ant Colony Optimization System

Abstract

This piece of research addresses an interesting comparative performance analysis and evaluation study for behavioral learning versus ant colony optimization. It considers two conceptual diverse algorithmic computational intelligence approaches. Both are related tightly to Neural and Non-Neural Systems respectively. The first algorithmic intelligent approach concerned with observed practically obtained results after one of neural animal systems’ activities. Namely, a mouse's active trials to reach an optimal solution for a reconstruction problem during its movement inside figure of eight (8) maze. Conversely, the second approach originated from realistic simulation results observed for Non-Neural system's activities namely: Ant Colony System (ACS). Obtained results have been reached while ACS is searching for an optimal solution of Traveling Sales-man Problem (TSP). Herein, some interesting observations have been introduced which concerned with similarity of enhancement for either learning systems under comparison. That enhanced/improved performance observed due to the effect of increase intelligent agent's number (either neurons or ants). Considering simulation of both adopted biological systems by Artificial Neural Network (ANN), results in very interesting findings. Furthermore, both have shown to be in agreement with learning convergence of an ANN learning model, while searching for optimal solution adopting Least Mean Square Error (LMS) Algorithm.