Neural networks and some practical applications in the field of artificial intelligence

Abstract

Even since Antiquity, in ancient Greece 4th century BC, or in China, the 3rd century BC, but not only there, people have imagined and even created automatic mechanisms, mechanisms that somehow were trying to be shy replicas of real examples of intelligent life from nature, including replicas of Man’s creation, such as the first bronze man, but nonetheless, non-intelligent replicas. Subsequently, in the modern age, attempts have been made to simulate a human neural network, as well as the first steps towards artificially simulating the functioning of the human brain. We will present in this article concepts of neural networks and some practical applications in the field of artificial intelligence and their capabilities to simulate autonomous thinking through advanced programming and complex assimilation of large data sets, using the three major learning paradigms, namely: supervised learning, unsupervised learning, and reinforcement learning. All these new concepts, over the past 75 years, and the advances in information technology have led to the emergence and development of practical applications of artificial intelligence. In this way, the human dream of creation tends to live and fulfil itself, why not, even in the coming years.