NOTE: Free essay sample provided on this page should be used for references or sample purposes only. The sample essay is available to anyone, so any direct quoting without mentioning the source will be considered plagiarism by schools, colleges and universities that use plagiarism detection software. To get a completely brand-new, plagiarism-free essay, please use our essay writing service.
One click instant price quote
Artificial Intelligence and Games Over the last 30 years, research in AI has fragmented into more and more specialized fields, working on more and more specialized problems, using more and more specialized algorithms. This approach has led to a long string of successes with important theoretical and practical advancements. However, these successes have made it easy for us to ignore our failure to make significant progress in building human-level AI systems. Human-level AI systems are the ones that you dreamed about when you first heard of AI: HAL from 2001, A Space Odyssey; DATA from Star Trek; or CP 30 and R 2 D 2 from Star Wars. (Atkin, 1999). They are smart enough to be both triumphant heroes and devious villains.
They seamlessly integrate all the human-level capabilities: real-time response, robustness, autonomous intelligent interaction with their environment, planning, communication with natural language, commonsense reasoning, creativity, and learning. (Laird, 2001). If this is our dream, why isn't any progress being made? Ironically, one of the major reasons that almost nobody (see Brooks et al. [ 2000 ] for one high-profile exception) is working on this grand goal of AI is that current applications of AI do not need full-blown human-level AI. For almost all applications, the generality and adaptability of human thought is not needed -- specialized, although more rigid and fragile, solutions are cheaper and easier to develop. (Atkin, 1999). Unfortunately, it is unclear whether the approaches that have been developed to solve specific problems are the right building blocks for creating human-level intelligence. The thesis is that artificial intelligence in interactive computer games is the killer application for human-level AI.
They are the application that will need human-level AI. Moreover, they can provide the environments for research on the right kinds of problem that lead to the type of incremental and integrative research needed to achieve human-level AI. (Laird, 2001). Bringing open architecture to a game represents a sea change for the game industry. Players are no longer bound to choosing a pre-set character in a battle game, or type of car in a race game. Instead, the game becomes as limitless as a player's creativity. This focus on the individual also affects how the game is played out on a virtual battlefield. (Wilson, 2001).
Games are ideal domains for exploring the capabilities of computational intelligence. The rules are fixed, the scope of the problem is constrained, and the interactions of the players are well defined. Contrast the game world with the real world -- the game of life -- where the rules often change, the scope of the problem is almost limitless, and the participants interact in an infinite number of ways. Games can be a microcosm of the real world (for example, the role of game theory in economics, social interaction, and animal behavior), and successfully achieving high computer performance in a nontrivial game can be a stepping stone toward solving more challenging real-world problems. (Schaeffer, 2001).
In a number of games, computers have enjoyed success that puts them on par or better with the best humans in the world. In some sense, these games are now the past, in that active research to develop high-performance programs for them is on the wane (or is now nonexistent). These include games where computers are better than all humans (checkers, Othello, Scrabble) and those where computers are competitive with the human world champion (backgammon, chess). Historically, games have been a popular choice for demonstrating new research ideas in AI. Indeed, one of the early goals of AI was to build a program capable of defeating the human world chess champion in a match. This challenge proved to be more difficult than was anticipated; the AI literature is replete with optimistic predictions.
It eventually took almost 50 years to complete the task -- a remarkably short time when one considers the software and hardware advances needed to make this amazing feat possible. Often overlooked, however, is that this result was also a testament to human abilities. Considering the formidable computing power that DEEP BLUE used in its 1997 exhibition match against world chess champion Garry Kasparov (machine: 200, 000, 000 chess positions a second; man: 2 a second), one can only admire the human champions for withstanding the technological onslaught for so long. Computer game research was started by some of the luminaries in computing science history. (Schaeffer, 2001).
In 1950, Claude Shannon published his seminal paper that laid out the framework for building high-performance game-playing programs. In 1951, Alan Turing (1953) did a hand simulation of his computer chess algorithm (a lack of resources prevented him from actually programming it); the algorithm lost to a weak human player. Around this time, Arthur Samuel began work on his famous checkers-playing program, the first program to achieve notable success against human opposition. By 1958, Alan Newell and Herb Simon had begun their investigations into chess, which eventually led to fundamental results for AI and cognitive science. (Schaeffer, 2001). "The nature of games played will change dramatically in the next few years. Turn-based games mandated by existing networks and devices will increasingly give way to visually satisfying action games, " said David Kerr, vice president of Strategy Analytics. "This will occur as manufacturers provide a model in which larger color displays are commonplace. " However, larger color displays will also require a new type of 3 G graphics chip that will run on mobile handsets. (Wilson, 2001). Given that our personal goal is to build human-level AI systems, we have struggled to find the right application for our research that requires the breadth, depth, and flexibility of human-level intelligence. (Wilson, 2001).
In 1991, the computer-generated forces were found. They were taken for large-scale distributed simulations as a potential application. Effective military training requires a complete battle space with tens if not hundreds or thousands of participants. The real world is too expensive and dangerous to use for continual training, and even simulation is prohibitively expensive and cumbersome when fully manned with humans.
The training of 4 pilots to fly an attack mission can require over 20 planes plus air controllers. The military does not even have a facility with 20 manned simulators, and if it did, the cost in personnel time for the other pilots and support personnel to train these four pilots would be astronomical. (Whatley, 1999). To bypass these costs, computer-generated forces are being developed to populate these simulations. These forces must integrate many of the capabilities we associate with human behavior -- after all, they are simulating human pilots. For example, they must use realistic models of multiple sensing modalities, encode and use large bodies of knowledge (military doctrine and tactics), perform their missions autonomously, coordinate their behavior, react quickly to changes in the environment, and dynamically reply missions. (Laird, 2001). In late 1997, we started to look for another application area, one where we could use what we learned from computer-generated forces and pursue further research on human-level intelligence.
We think we have found it in interactive computer games. For a number of games, a short history of the progress in building a world-class program for the game is given, along with a brief description of the strongest program. In each case, a single feature of the program that is a major contributor to the program's strength is highlighted. The biggest advances in computer game playing have come as a result of work done on the alpha-beta search algorithm. This algorithm received the most attention because of the research community's preoccupation with chess.
With the DEEP BLUE victory over world chess champion Garry Kasparov, interest in methods suitable for chess has waned and been replaced by activity in other games. One could argue that the chess victory removed a ball and shackle that was stifling the creativity of researchers who were building high-performance game-playing systems. (Time, 1995). The alpha-beta research led to a plethora of search enhancements, which significantly improved the efficiency of the search. Some of these enhancements include iterative deepening, caching previously seen subtree results (transposition tables), successor reordering, search extensions and reductions, probabilistic cutoffs, and parallel search. The results are truly amazing.
Even though there is an exponential difference between the best case and the worst case for an alpha-beta search, most high-performance game-playing programs are searching within a small constant of the best case. (Whatley, 1999). With a potential one billion people watching, 27 venues to cover and an estimated nine million printouts for the 17 th Commonwealth Games, IT cannot afford to fail. While Microsoft and XML are new to this game, they will underpin the event's IT. (Mactaggart, 2002). It is almost an axiom of IT projects that deadlines are elastic. Poor planning or poor project control may be major factors but large and complex IT projects very often run late or over budget, or launch with reduced specification and inadequately tested. For Gerry Pennell, IT director of Manchester 2002, the company formed from a consortium of local authorities and commercial companies to manage the 17 th Commonwealth Games, having an absolute and fixed deadline of 25 July concentrates minds wonderfully. (Mactaggart, 2002).
Artificial intelligence has always joined mundane goals with lofty ones. On one level the field has been about making clever devices: computers that understand English, systems that diagnose diseases, toys that play games. On another level, though, such devices are supposed to serve as windows on the mind. AI investigators have long maintained that intelligence can be understood strictly in terms of computation -- and apart from any physical systems that happen to exhibit it.
By dissecting and characterizing intelligence, then building its components into devices that can do some of the things that people do, engineers and programmers hope to succeed where philosophers and psychologists have failed: they hope to develop a scientific understanding of intelligence. (Mechner, 1998). In the late 1950 s, when computer programs began, however tentatively, to approximate elements of human reasoning, AI investigators made some bold predictions. Within a decade, many said, computers would be as intelligent as humans: they would communicate with people in English and discover important mathematical proofs. But distilling the calculus of cognition proved much harder than most people expected. (Mechner, 1998). Intelligence demands knowledge, of course, but most of what people know is so deeply ingrained that it hardly seems like knowledge at all: that a cup of water, tipped over, will spread onto a table; that painting a wall will not change the shape of the wall or cause a flash flood in Texas.
But those bits of common sense are embedded in a nervous system that has been optimized over millions of years of evolution. Because such common sense is still impossible to reproduce in a computer, programs designed to deal with some part of the real world "intelligently" can do so only by focusing on one thing and ignoring everything else. A program called MYCIN, for instance, can correctly diagnose a patient with meningitis if certain symptoms are described to it. But...
Free research essays on topics related to: level intelligence, artificial intelligence, common sense, computer generated, deep blue
Research essay sample on Artificial Intelligence Level Intelligence
