Wordtrade.com
Technology

Gaming Design

Biologically Inspired Artificial Intelligence for Computer Games by Darryl Charles, Colin Fyfe, Daniel Livingstone, Stephen Mcglinchey (IGI Publishing) Computer games are often played by a human player against an artificial intelligence software entity. In order to truly respond in a human-like manner, the artificial intelligence in games must be adaptive, or respond as a human player would as he/she learns to play a game. Biologically Inspired Artificial Intelligence for Computer Games reviews several strands of modern artificial intelligence, including supervised and unsupervised artificial neural networks; evolutionary algorithms; artificial immune systems, swarms, and shows using case studies for each to display how they may be applied to computer games. This book spans the divide which currently exists between the academic research community working with advanced artificial intelligence techniques and the games programming community which must create and release new, robust, and interesting games on strict deadlines, thereby creating an invaluable collection supporting both technological research and the gaming industry.

Why would anyone want to make a computer game intelligent? What motivation is there for putting intelligence into perfectly good games? After all, humans have played games since prehistory and we have never had the ability to create intelligence in our games until now. This, of course, overlooks the fact that the authors typically played games against other human beings (most board games and sports) or against ourselves (often memory based games), sometimes with a statistical element involved (such as the solo card games of `patience'). Most people would agree that the most satisfying is playing games against other people, particularly when the other people are of the same standard as ourselves. There is something inherently satisfying about competition with our peers. Adults do not get the same satisfaction competing against, for example, children who do not yet have our capabilities. Similarly most get no fun at all playing against an opponent who is perceived to be much better than us at a particular game. Thus we want to create games in which the opponent is, in some way, a worthy opponent for us. The computer promises to liberate us from the need to find such opponents by providing us with artificial opponents of a suitable ability. Thus our computer games commonly come not just with the game itself but with built-in artificial opponents who aspire to being worthy opponents for us.

Even in the realm of online multiplayer gaming, nonplayer characters (N PCs) are a required staple of many games, such as in most massively-multiplayer online games (MMOG), where N PCs may be required to provide missions or combat encounters. Here the contrast between intelligent players and nonintelligent computer controlled characters is even sharper, and clearly an area where more machine intelligence could be applied to improve the game for all the players.

This book is concerned with the application of modern artificial intelligence (AI) techniques to computer games. The authors suggest that the current generation of computer games is largely free of this technology. You may well protest that Al is already well established in such games; after all the computer opponent in such games not always known as the 'Al.' But the authors highlight the use of the word 'modern.' Most, though by no means all, games of recent and current generations use fairly old and certainly rather standard forms of artificial intelligence. In this book the authors concentrate on technologies which are still the subject of large scale effort in the research field, such as artificial neural networks, genetic algorithms, or artificial immune systems. The authors shall not discuss, for example, case based reasoning since they concentrate solely on techniques which are based on those thought to dominate in real life.

So what is it that these technologies bring to computer games that finite state machines, expert systems, and the rest do not bring? The unifying theme of these techniques is adaptation, whether as a means to learn how to play a game or even while the game is being played. In artificial neural networks the authors are changing the network weights—corresponding to the synaptic efficiencies—so that the response from the artificial neural network changes from game to game. In evolutionary algorithms, the artificial chromosomes are being selected base on their fitness so that better solutions are being found as the game proceeds; in artifi­cial immune systems, our artificial antibodies are being produced in response to attacks by artificial antigens which allows the artificial immune system to adapt to varying threats.

Why do the authors consider adaptation to be so central to creating intelligence in computer games? We feel that most people playing against an Al wish to feel that they are playing against another intelligent person. When two people play a competitive game, each uses a continu­ally changing strategy: if you know your opponent's strategy in advance, you can prepare a defence specific to such a strategy. Even in completely nonrandom games such as chess, no player will use exactly the same opening or defence in each game. To do so would leave an opponent with a very simplified problem and would certainly result in long-term defeat. This is not to say that we will not have a set of preferred strategies in a game but the emphasis must be on having a nonsingular set, not just a single strategy.

There is one other point which may be made with respect to these biologically inspired Al techniques. Competition (for food, for shelter, for mates) is built into life itself and so the authors consider that using analogies in software of real biological processes means that the authors are basing our techniques on methodologies which have already proved successful in real life. Thus the authors may anticipate that the resulting techniques match in some way the experiences the authors already have as human beings in our existing environments.

By focussing on biologically-inspired AI techniques, rather than just those methods com­monly used in the games industry, some sections of this book may appear to be more theo­retical than applied. A number of case-studies are scattered throughout—either as chapters in their own right, or included into larger chapters—to offset this. These vary from simple illustrative examples, to case studies in contemporary action and racing games.

This book arose from their experience of teaching on a range of modules related to computer game development, including modules specifically focussed on game Al. The authors found that there was no existing text which satisfied their desired pedagogical aims. Many textbooks on game Al tend to be devoted to state space methods and efficient searches. Most present algorithms and solutions, but in concentrating on showing how to achieve results, they provide little in the way of theoretical underpinnings. A notable exception is that field which is devoted to evolutionary game theory; however, this field is much more based on mathematical game theory which is outside the interests of most programmers who are investigating computer games.

The research background of several of the authors is in the broad area of artificial neural networks (ANN), and this accordingly features in several of the chapters. It was their inten­tion, however, to present a book which presented a more complete, yet still academic, view

of game Al. The authors have also been able to draw upon their experience from research in a range of related areas, from artificial life and genetic algorithms to artificial immune systems and ant-colony optimisation.

In Chapter I, the authors present a brief review the history of computer games, from the I 950s to the present day. Needless to say the rate of change over the last decade has been much greater than that seen in earlier times and this acceleration seems likely to continue but the authors have tried to do justice to the major trends in existence today. The goal of this chapter is to more clearly set out what the common goals of game AI are, and to explain why much of modern game Al is not of academic interest.

In Chapter II, the authors provide a broad introduction to natural and artificial neural networks, including notes on the history of the development of modern artificial neural networks. Then, in Chapter III, the authors introduce the technique which, to many people, is artificial neural networks, the multilayered perceptron, or MLP. This well known technique uses the back-propagation of errors to learn. The authors give the theory behind the learning and some heuristics about its performance before illustrating its use on some general data problems. The authors then illustrate how it may be used in a game context. In Chapter IV, the authors provide a short case study on the use of artificial neural networks—particularly the multilayered perceptron—in digital games, using the Robocode competition framework as a suitable example.

The artificial neural networks in Chapters III and IV use supervised learning to adapt; the training set must include a set of answers appropriate to the problems in hand. However, a lot of human and animal learning occurs without a teacher. In particular, when playing a game we may learn by creating internal models of the game as it is being played, without a teacher taking on the role of a 'learning supervisor.' Thus in Chapter V, the authors discuss two types of artificial neural networks which organise according to unsupervised learning principles. We differentiate between competitive learning, a variation of which leads to self-organising maps, and Hebbian learning; the former tends to be used for clustering a data set while the latter leads to low dimensional projections of a data set. The self-organising map is used to develop AI opponents in the rather old game of Pong while the projection methods are illustrated in a high dimensional hide-and-seek game.

The authors stated that, for many people, artificial neural networks are equated with the multilayered perceptron, however the multilayered perceptron is actually only one network in a rather large field of supervised learning methods. Therefore in Chapter VI, the authors introduce a further alternative which can be used instead of the multilayered perceptron. The authors pay particular emphasis on the radial basis network since its speed of learning makes it extremely attrac­tive in the context of computer games. Returning to Pong, the authors are then able to compare the in-game performance of different artificial neural networks in creating players for this simple game.

In Chapter VII, the authors switch to evolutionary algorithms, the most famous of which is the ge­netic algorithm. The authors spend some time examining issues of representation before illustrating its use on a computer game. The case study presented in this chapter, a motocross game with realistic physics, is a particularly challenging problem for Al to solve; the solution presented combining the power of both evolutionary and artificial neural net approaches. The games industry's traditional solution to this problem has been to use designer intel­ligence rather than machine intelligence and have the developers implement the paths to be followed. This work clearly demonstrates that contemporary games can potentially benefit from biologically inspired Al.

In Chapter VIII the authors explore just some of the many ways in which the standard genetic algorithm can be extended, and look at a few alternative evolutionary methods. Deceptive problems are introduced before the authors consider the first extension of the structured GA. Other alternatives introduced include probability based incremental learning (PBIL), which is also demonstrated through an example of the N-Persons Iterated Prisoners' Dilemma.

Chapter XI introduces the basic concept of multiobjective problems, with particular relation to the genetic algorithm. It then considers how genetic algorithms and evolutionary methods might be applied in the real-time strategy genre of computer games. In such games, Al has to work at a number of distinct levels, roughly speaking covering the strategic goals and the low level tactics. Exploring the full set of strategies and finding all the good solutions in such situations is far from straightforward, and this chapter illustrates some of the issues.

In Chapter X, the authors discuss artificial immune systems and show how their resulting models can be used in a variety of computer games. Since this topic is rather newer than the other Al techniques, the authors spend some time developing the artificial immune systems in a variety of game situations.

Two further, very distinct, biologically inspired Al techniques are introduced in the fol­lowing two chapters. In Chapter XI ant-colony optimisation (ACO) is demonstrated as an alternative method of path-planning. An example application, the classic game of Combat is presented.

The following chapter, Chapter XII, outlines the methods used in reinforcement learning. Although the term 'reinforcement learning' can be used more generally to describe the learn­ing methods used in a variety of Al techniques, it is also the name of a particular approach and set of algorithms for Al learning. The reinforcement learning approach is one based on the payment of rewards for actions leading to good results, such as winning a game. This simple notion forms the basis for a group of methods which at their core resemble human game playing approaches in a number of ways, although which have yet to be used to any extent in the commercial world of game development.

The authors close with two chapters which focus less on particular Al methods, but on issues relating to their use and misuse. Chapter XIII discusses adaptivity in games more generally. Cur­rent approaches to adaptivity in commercial games are discussed, along with issues facing developers wishing to introduce and develop adaptive solutions. It is their feeling that most attempts to include adaptivity in games have been (often deliberately) limited, and that there is a great deal of potential in making games more adaptive in a range of areas.

Finally in Chapter XIV, the authors assess how successful their attempt to install human-like intel­ligence in computer games has been using methods suggested by the famous Turing Test. It is of interest that Turing expected computers to pass his test by the year 2000. The authors believe that success in the game version of the Turing Test is more reachable in the near future than success in the standard Turing Test.

The book need not be read in any particular order and the chapters are intended to be self-contained, the exception being that Chapter III should be read before any of the other chap­ters on artificial neural networks. Thus the authors have also provided references at the end of each separate chapter rather than a single section at the end of the book.

Quests: Design, Theory, and History in Games and Narratives by Jeffrey Howard (A K Peters Ltd) This book is intended for a broad range of audiences, all of whom can take something useful from each of its sections.

• As a bridge-building text, it belongs in the toolkits of both humanities scholars and independent designers.
• New media researchers can benefit from the analysis and modi­fication of the theories of quest narratives.

Humanities scholars and professors, such as literature teachers wanting to bring computer-assisted instruction into their class­room in an innovative way, can benefit from the book's combination of literature, games, and practical classroom exercises. Such professors might consider using this book in a class on the relationship between narratives and games, where it would work well as a practical and accessible textbook. This book includes many tutorials and exercises for use with the Aurora Toolset, which can be purchased with the role-playing game Neverwinter Nights for fifteen dollars or less.

Independent designers seeking to create new, innovative quest games can benefit from the theoretical arguments and tutori­als, which present useful skills in the accessible, user-friendly Aurora toolset.

Because the audiences for the book are varied, I have tried to keep overly obscure theory to a minimum, relegating postmodern ideas about language and interpretation primarily to footnotes. Similarly, I have provided histori­cal background on both quest games and quest narratives, assuming that a reader who knows the acronyms for role-playing terminology may not know the plot of Sir Gawain and the Green Knight and vice versa. When designers, scholars, and students share knowledge of their respective disciplines and col­laborate on focused, unified projects, the results are likely to help everyone involved.

After a general introduction to the theory and history of quests, this book contains four chapters about four aspects of quests, each consisting of two subsections. The chapters discuss four theoretical components of quests:

• spaces;
• objects;
• actors;
• challenges.

For each theoretical subsection, there is a corresponding section describing a practical skill associated with this aspect of the quest, with accompanying exercises and suggestions for the use of particular technologies for designing aspects of quests. These four practical sections are:

• level design; quest-item creation;
• NPC and dialogue construction;
• event-based programming, or "scripting."

Understanding the theory of the spaces of the quest can improve the prac­tice of level design, and understanding the role of objects in quest narra­tives and games can help one to craft interesting treasures as goals or rewards of one's quests. Studying NPCs leads to better quest dialogue, and know­ing the central challenges of quests leads to better gameplay in the form of scripting.

These correspondences are deliberate, since a key purpose of the book is to build bridges between the theory of literature and games and the practice of game design, both independently and commercially. There is a tendency on the part of many game designers to view theory skeptically because they consider it to be divorced from practice, like Chris Crawford's diagram of a fractured pyramid in which a base of "Interactive Storytelling" is separated from a broken-off tip contemptuously labeled "AArseth [sic] et all [sic]" (74). Crawford's argument is that game designers often do not think that theory can help them to make a better product—an accusation that is frequently leveled against theorists of the ludology/narratology debate. Yet I would re­spond that designers do not benefit from ignoring the theory or history of quest games or quest narratives, and they may even lose out significantly by being forced to reinvent the wheel of quest design, struggling with a design problem that has already been solved in the history of quest games or quest narratives.

Some game designers might suggest that understanding the "meaning" of quests is not important to design, but this objection overlooks a funda­mental aspect of quest design: the audience of role-playing games and adven­ture games. A player who goes to a game store and selects a role-playing game or an adventure game over a shooter or a simulation is likely to already have an investment in an epic experience. Supporting this idea, Chris Bateman in 21" Century Game Design offers an audience-centered approach to game design that draws upon extensive quantitative research of audience prefer­ences to classify marketable games, including three overall genres categories of "quest," "strategy," and "simulation." "Quest" games include the genres of adventure and role-playing games, suggesting that these two genres are linked by a shared central activity. As Bateman argues, the term quest accurately describes the core value of the games grouped here. All tell stories, and, due to the nature of the medium, these stories tend towards the epic (with more intimate stories better suiting action games). Players expect their quest games to last many hours of play, so stories are often wide in scope. (264)

From Bateman's perspective as a successful designer who has done substan­tial research into self-reported audience preferences, many gamers prefer a sweeping, epic experience that can best be described as a "quest game," which includes a variety of more commonly used genre labels such as "adventure game" or "RPG." One example of a quest-based RPG that has benefited from its designers' consideration of the meaning of quests is Neverwinter Nights 2. The tagline "everything you do has a meaning" points toward the relevance of meaningful action in a commercially successful RPG. This is a deliberate marketing strategy geared toward the idea that players of quest games buy their games on the basis of meaningful quests.2 Such players want a game to give them a sense that their play is part of an epic storyline with consequences that will affect a simulated world positively or adversely. These players will continue to appreciate their favorite quest forms, whether kill quests or fetch quests, but they will also gravitate toward games that use these forms or oth­ers in new and interesting ways.

There are at least three different ways in which quests can be meaning­ful, with several subcategories and ways in which these types of meaning can intersect and complement each other.

Meaning can refer to:

• the impact of the player's accomplishments on and within a simulated world, including
• achieving greater power in this world and understanding of its lore, in gaming terminology as "leveling up" and in mythological terms as "initiation";
• changing the shape of the landscape;
• altering the political and moral balance of the game world; changing the player-avatar's relationship with her compan­ions or other NPCs;
• a narrative backstory that conveys emotional urgency by reveal­ing why the player-avatar is performing an action and what ef­fects this action will have;
• expressive, semantic, and thematic meaning: ideas symbolically encoded within the landscape, objects, and challenges of the quest and enacted through it.

Players' uncovering of thematic meaning through the completion of a quest can sometimes entail literally deciphering a secret message in a puzzle. More often, however, the world itself is the puzzle. The ways that different players go about completing quests "reconfigures" these meanings, allowing players to experience them in many ways.

This enactment of meaning is potential and metaphorical in quest narratives but actual in quest games. In narratives about quests, characters perform deeds that have meaning, often allegorical. For example, The Quest for the Holy Grail follows a pattern in which a knight's performance of a series of complex actions is followed by an explanation of the spiritual significance of each part of these actions by a wise figure, such as a monk. This explana­tion is usually complex yet blunt in its delivery, as when a monk says to Sir Galahad, "Sir, you asked me just now the meaning of the mysterious task you fulfilled, and I will gladly inform you" (63). In a postmodern narrative, these meanings might change according to the interpretation of the reader. If the plot is significantly ambiguous, a reader's interpretation may also af­fect the imagined "outcome" of these events in her mind. However, in both postmodern and premodern narratives, the characters' actions for the most part remain the same from reading to reading. Moreover, the reader does not have to exert effort to cause these events to occur because they will be narrated no matter what. The reader exerts cognitive and interpretative effort to determine what the actions performed by a character mean after they have occurred. In contrast, players of quest games must sometimes determine what their actions mean in order to perform them correctly. Their meaning is discovered or created by a player through action, so that the action itself is meaningful. It is not just the events that are "realized" or "actualized," in Juul's terms, but the meanings of these events as well.

For example, Oblivion is thematically "about" the struggle of good and evil and the necessity to act heroically to fend off destruction of self and soci­ety, although this theme must be actively pursued by the player in order for it to communicate. Thus, if the player chooses primarily to wander the world of Tamriel seeking buried treasure in dungeons, she would avoid exposure to this theme but would miss large portions of the actions available in the game. The main quest is emotionally and morally significant because it concerns the welfare of an entire simulated world, which the player "cares" about more strongly than a fictional world under attack because it is her actions that can save it. The main quest establishes its emotional significance in the opening stages of the game, in which the player must protect the emperor. The em­peror, dramatically voiced by Patrick Stewart, issues a "call to adventure" by bravely facing a death prophesied in ominous dreams and rescuing the player character from jail so that he can save Tamriel. The side-quests elicit similar forms of emotional investment through character development and dramatic voice acting, leading the player to care about an unfortunate NPC who has lost a valuable item or close friend.

Meaning as Initiation

Quests are meaningful because they immerse players in dramas of initiation, defined as a gradual movement up through formalized "levels" of achievement into a progressively greater understanding of the rules and narrative in a simulated world. Initiation also entails insight into how this world com­ments imaginatively upon "real" events, circumstances, and ideas. Designers create an activity out of the quest structure by lengthening and repeating the phase of initiation described by Campbell, which originates in the trade guilds of the middle ages and the mystery religions of ancient Greece and Rome. The impetus behind quests is achievement, motivated by a drive to overcome difficulties and to better oneself, both in the virtual form of one's "avatar" and in the real-world skills developed through extended play­ing. What Frye glosses over as the "minor adventures" and "perilous journey" leading up to the hero's final battle becomes the center of quest in games, because these episodes are the greatest source for action (187). As Campbell writes, the hero "must survive a succession of trials. This a favorite phase of the myth-adventure. It has produced a world literature of miraculous tests and ordeals" (97). While the Campbellian elements of the initiation appear frequently in games, Oblivion reaches far back into the history and structure of initiation in order to produce compelling gameplay. There is a set of "quest lines" called "guild quests" related to the player character's membership and status within "guilds" devoted to particular professions in the game, such as a "mages' guild" of wizards, a "fighters' guild" of warriors and knights, as well as a shadowy "thieves' guild" of assassins and criminals. In massively multi-player online role-playing games (MMORPGs), these guilds facilitate com­munity formation by offering support and a sense of belonging. At the same time, these organizations are based upon the trade guilds of the middle ages, which measured and validated a craftsman's increasing abilities. Hence, the "skill levels" associated with particular abilities like combat or alchemy derive directly from medieval guild vocabulary, including "novice," "apprentice," "journeyman," "expert," and "master." In a role-playing game, a character's abilities are measured by his "level," representing a summary evaluation of his experience within the game and his resulting abilities. Much of the motiva­tion for extended play, which can take hours of mental exertion, comes from the desire to "level up" by gaining sufficient experience to ascend to the next level. The idea of achieving greater "levels" within an organization based upon skill and understanding comes from trade guilds, but also from the mystery religions and secret societies that preceded them and developed out of them, such as the Eleusinian mysteries or the Freemasons. For example, in the course of playing over many weeks or months, a player might attempt to move his character from a "first level" Spellsword to a "thirteenth level" Spellsword. This degree of formalized ascent through numbered gradations that bring new abilities and insights resembles the process of a "neophyte" of a mystical order becoming an "adept" or a new initiate into Masonry attempt­ing to become a thirty-third degree Freemason."

Meaning as Narrative

While the primary type of meaning in quests is achievement in the form of initiation, a second form of meaning is derived from narrative, which motivates the player through a back story that gives urgency to a task, or rewards the player through an explanation of the events that occur as a result of the task's completion. A successfully completed quest may reveal what happened to lead up to it, such as "whodunit" in a murder mystery, or what will happen as a result of it, such as the fate of the empire after the demonic antagonist has been vanquished. While quests are not synonymous with narratives, the drive to discover what happened before or what will happen after the quest can be a powerful motivation to invest large amounts of time and effort in a game.

Thematic Meaning

A third form of meaning is thematic meaning, communicated when the player acts out a set of ideas that comment upon the simulated world of the game and the "real" world outside of it. This commentary can run the gamut from simple themes, such as a basic conflict between good and evil enacted in the game's battles, to sophisticated puzzles that conceal encoded meanings through the interrelationship of their parts. As in any other work of art, the themes enacted by players can fall into a wide variety of categories, including natural, psychological, political, and religious ideas. Designers can create quests with thematic meaning by drawing upon a tradition of symbolic cor­respondences that runs throughout quest narratives and through some of the most sophisticated quest games.

 

Headline 3

insert content here

WT Main | About WT | Review Links | Contact | Review Sources | Search