The Role and Place of Artificial Neural Network Architectures Structural Redundancy in the Input Data Prototypes and Generalization Development  

作　　者：Conrad Onésime Oboulhas Tsahat Ngoulou-A-Ndzeli  Béranger Destin Ossibi Conrad Onésime Oboulhas Tsahat;Ngoulou-A-Ndzeli ;Béranger Destin Ossibi(Ecole Nationale Suprieure Polytechnique, Universit Marien Ngouabi, Brazzaville, Republic of Congo)

机构地区：[1]Ecole Nationale Suprieure Polytechnique, Universit Marien Ngouabi, Brazzaville, Republic of Congo

出　　处：《Journal of Computer and Communications》2024年第7期1-11,共11页电脑和通信（英文）

摘　　要：Neural Networks (NN) are the functional unit of Deep Learning and are known to mimic the behavior of the human brain to solve complex data-driven problems. Whenever we train our own neural networks, we need to take care of something called the generalization of the neural network. The performance of Artificial Neural Networks (ANN) mostly depends upon its generalization capability. In this paper, we propose an innovative approach to enhance the generalization capability of artificial neural networks (ANN) using structural redundancy. A novel perspective on handling input data prototypes and their impact on the development of generalization, which could improve to ANN architectures accuracy and reliability is described.Neural Networks (NN) are the functional unit of Deep Learning and are known to mimic the behavior of the human brain to solve complex data-driven problems. Whenever we train our own neural networks, we need to take care of something called the generalization of the neural network. The performance of Artificial Neural Networks (ANN) mostly depends upon its generalization capability. In this paper, we propose an innovative approach to enhance the generalization capability of artificial neural networks (ANN) using structural redundancy. A novel perspective on handling input data prototypes and their impact on the development of generalization, which could improve to ANN architectures accuracy and reliability is described.

关 键 词：Multilayer Neural Network Multidimensional Nonlinear Interpolation Generalization by Similarity Artificial Intelligence Prototype Development 

分 类 号：TP1[自动化与计算机技术—控制理论与控制工程]