“A Review on Game Playing in AI” Name: Sagar Sanjay Khomane Student: VPASC College, Baramati. Email:sagarskhomane14@gmail.com
Abstract:
Games hold an inexplicable fascination for many people and the notion that computers might place games has existed at least as long as computers. The ideal way to use a search procedure to find a solution to a problem is to generate moves through the problem space until a goal state is reached. In the context of game-playing programs, a goal state is one in which we win. In the New AI Challenge and Competition, the general game playing has quickly evolved into an established research area.…show more content… Unlike specialized systems like the chess program, Deep Blue, a general game player cannot rely on algorithms signed in advance for specific games. Such a system rather requires a form of general intelligence that enables it to autonomously adapt to new and possibly radically different environments. General game-playing systems are a quite essential example of a new generation of software that end users can customize for their own specific tasks. This makes general game playing an interesting and challenging problem for AI, involving many fundamental issues such as reasoning, learning, planning and decision making. Consequently, general game playing can, and in fact should, be of interest to researchers in a variety of AI disciplines beyond conventional computer game playing. At the same time and for the same reasons, general game playing provides a new anchor for AI education as a unique and attractive for student’s framework for teaching multiple basic AI topics, such as problem solving by search, propositional and order logic, logic programming and planning. AI researcher does not lie in general game-playing systems themselves, they can be used as a non-trivial application for a broad range of more theoretically motivated AI research. A further advantage of general game…show more content… Other areas are likely to eventually become more relevant as the scope of general game playing is broadened decision making. Current state-of-the-art research in general game playing focuses on games that are deterministic and in which players have complete knowledge of the state. As we progress towards the general case of randomized, imperfect-information games, aspects of general decision making will become increasingly important, including utility functions, Markov decision processes, and game-theoretic solutions. In Natural Language Processing most end users of general game-playing systems will expect a better way than using GDL to input the rules of their own invented games. This could be achieved with the help of some controlled form of natural language for describing games. In most cases, this will include a process of disambiguation through a dialogue between the player and its user. This generalization seems particularly attractive because it could be met in a mostly modular fashion, where an existing general game player is coupled with a pre-processing natural language interface game playing robots. General game-playing software provides a great opportunity to make a relatively simple robotic system to act