Διάσχιση λαβυρίνθου με την χρήση τεχνητών νευρωνικών δικτύων
Maze solving with artificial neural networks
KeywordsΤεχνητά νευρωνικά δίκτυα ; Αλγόριθμος της οπισθοδιάδοσης ; Εφαρμογές τεχνητών νευρωνικών δικτύων
This dissertation provides a broad description of artificial neural networks, a science that has grown significantly in recent years. The topics fall under the area of Artificial Intelligence, which can find solutions to many problems in various fields such as health, economics, meteorology, pattern recognition and transportation. and is expected to play a key role in the technology of the future. Artificial Intelligence is the creation of machines that mimic human behavior. This paper presents the use of neural networks for the problem of labyrinth crossing. Particularly : The first chapter describes biological neurons and biological neural networks. This is important because artificial neural networks are biologically inspired. The second chapter includes an extensive analysis of the artificial neuron, the artificial neural networks. It also analyzes the similarities, and differences, between artificial neural networks and biological neural networks and presents the advantages and disadvantages of artificial neural networks. The third chapter describes the process of training artificial neural networks with the regression propagation algorithm. The fourth chapter describes the crossing of a maze using an artificial neural network. The fifth chapter describes areas in which artificial neural networks have found great application.