A Comparative Evaluation of Neural Network Architectures and Genetic Algorithms for Adaptive Decision Making in Autonomous Systems

Abstract

In the pursuit of enhancing adaptive decision-making capabilities in autonomous systems, this study presents a comparative analysis of various neural network architectures and genetic algorithms. We systematically evaluate the performance of convolutional neural networks (CNN) in conjunction with genetic algorithm optimization against traditional multi-layer perceptron (MLP) configurations. Employing a series of simulations across diverse autonomous scenarios, we measure efficiency, accuracy, and adaptability metrics. The results indicate that hybrid models combining CNNs with genetic algorithms significantly outperform conventional MLPs, particularly in complex, dynamic environments. This research not only contributes to the theoretical framework surrounding adaptive systems but also offers practical insights for future AI developments in autonomous technologies.