Development of Artificial Intelligence: An Approach using Alpha-Beta Pruning

Abstract

Over the fifty years that artificial intelligence has been created and defined, there have been many people who have wrote on and discussed the topic. However, the field of artificial intelligence is extraordinarily broad and difficult to encapsulate thematically as there are so many forms of artificial intelligence. Artificial Intelligence by definition is “the theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages”, through this definition we can understand that there are many different forms of artificial intelligence. The most basic forms (and the form of artificial intelligence that I have researched fluidly and will also be discussing) are what we call “Reactive Machines”. We call them “Reactive Machines” as they are purely reactive, and do not have the ability to form memories nor do they have the ability to use past experiences to form a decision based on the current situation. They have no concept of the past and they also lack any memory of past events. Because of the nature of these so called “Reactive Machines” they are forced into only making predictions about the future events and choose the most optimal thing to do based on the situation at hand. These qualities allow Reactive Machines to play specific video games very well as they are not easily fooled, with the only downfall of these forms of artificial intelligence being that they are incapable of knowing anything other than what they are programmed to know. In this paper I shall discuss the creation process of a Chess Reactive Machine and the algorithm that was used in the process.