چکیده مقاله
In this article, we extensively examine and evaluate a neural network with a full connection architecture This neural network constitutes a complex computational model that, through establishing connections among all neurons across different layers, possesses the capability to learn and generalize intricate patterns and complex features The article commences by introducing fundamental concepts of neural networks and the full connection architecture Subsequently, we meticulously delineate the stages of neural network analysis and evaluation, encompassing data aggregation, network model selection, determination of layer and neuron quantities, specification of activation and cost functions, execution of training with training data, assessment with test data, and optimization of weights and biases Ultimately, by presenting a practical example and simulating it using MATLAB, we engage in a more meticulous investigation of the neural network with a full connection architecture This simulation empowers us to delve into a more precise analysis of the network's performance and operations in reality, allowing us to evaluate the outcomes attained The precise analysis and comprehensive evaluation of a neural network with a full connection architecture play a crucial role in the development and enhancement of artificial intelligence technologies This article endeavors to provide accurate information and analytical perspectives, aiming to foster a deeper comprehension of this intricate architecture and contribute to the facilitation of developments in AI related domains
نویسندگان
Alireza Norouzpour Shahrbejari
Department of Electrical EngineeringTechnical and Vocational University TVU Tehran, Iran
Morteza Ghorbani
Department of Engineering, School of Mechanical Engineering, College of Engineering, Islamic Azad university of Mashhad, Mashhad
Amirhossein Pourashraf
Islamic Azad University Isfahan Khorasgan Branch
شیوه ارجاع
Norouzpour Shahrbejari, Alireza and Ghorbani, Morteza and Pourashraf, Amirhossein,1402,Analysis of a Neural Network with a Propagation Architecture,The 16th International Conference on Innovation and Research in Engineering Sciences
ارائهشده در
مجموعه مقالات شانزدهمین کنفرانس بین المللی نوآوری و تحقیق در علوم مهندسی19 آبان 1402