 Introduction to aLife Interactive Guide The aLife Interactive Essay is a compilation of knowledge by Stewart Dean on the topic of artificial life. Mr. Dean has integrated the wisdom that he acquired through self-education, online information, and various books into an easy-to-understand collection of articles on the many facets of artificial life (also known as aLife). Mr. Dean?s purpose in compiling the acquired learning into an interactive essay was to demonstrate that "Life is within our understanding and can be, to various levels, recreated in computer environments." He goes on to poetically state that in understanding aLife, "we are not inventing anything, just discovering what nature and the universe has always known." I found Mr. Dean?s observation to be beautifully stated and logical. Sometimes when "new" scientific information is reported, the audience is led into believing that the findings are novel and previously unknown. In reality, however, the data might have been common knowledge to individuals in the particular field but not widely recounted to the public. On the hand other, as Mr. Dean stated, it might be a known phenomena replicated in an unconventional manner. The sections which Mr. Dean addresses in his interactive essay are as shown in the table below:
After scanning and/or reading thousands of web pages devoted to the general areas of artificial intelligence (AI) and artificial life (aLife), I have found the aLife Interactive Essay to be a superior site in terms of its structure and wording. The vast majority of web documents on AI and aLife assume a general familiarity with programming and a specific emphasis on AI programming. Although I have been personally fortunate to have experience in these areas, the technical tone of these sites did not explore the essential basics. In my opinion, these basics would include a brief general introduction to aLife, with expansion on various aspects of aLife such as emergence, entropy, chaos theory, creatures, neural networks, and evolution.. I was also immensely pleased with Mr. Dean?s section on The Uses and Abuses of aLife because with a powerful "tool" such as aLife, it is important to be aware of both it?s strengths and weaknesses. Desiring to locate a web site which could appeal to individuals possessing a broad range in level of exposure to and experience with computers, I was extremely pleased to find Mr. Dean?s site. This site was configured in a style which would allow an individual to either learn about an introduction to the entire scope of aLife, or just concentrate on a particular portion depending on the level of expertise that s/he possessed on this topic.
|
  Article 1: An Introduction to aLife Artificial Life or aLife describes the field which resulted from the merging of the endeavors of various disciplines including behavioral biologists and computer scientists. As an illustration, the behavioral biologists who sought to understand behaviors of nature collaborated with the computer scientists who wished to create programs which would mimic the conduct of nature?s creatures. Although initially the computer scientists were the primary beneficiaries of the shared knowledge, the biologists also profited from the increased proficiency of the computer scientists. The models of behavior provided by the biologists were used to create programs which followed the logical paths that would be taken in nature. As the computer programs began to "learn" from the actions of previous "generations", the subsequent "behaviors" exhibited by the programs were relayed to the biologists who could then seek the "new" patterns in nature. Mr. Dean provides two definitions for life. The first from his Oxford dictionary is as follows: Mr. Dean also stated in his article that "Nature could be said to be one application of the rules of life." Although a simple statement, I find that it is very profound in its simplicity. Even on our own planet, in our own homes, there are life forms which we may not regularly acknowledge. As an example, each home to some degree possesses colonies of dust mites. These colonies are not visible to the naked human eye, however the life forms do exist. In the same manner, possibly there are other forms of life that exist both on our planet and elsewhere which are not currently perceptible to us. Even if we?re unable to perceive them, they still exist. In the case of aLife, as Mr. Dean eloquently says, "Alife is only artificial in the sense that is it man made." From a theoretical standpoint, who are we to say that the aLife creations are not alive just because they were programmed by a human being? The aLife systems evolve and adapt to their environments as do other living creatures. Yes, the aLife programs obtain their initial form and "life instructions" from humans, but each living being also at one point in time (long, long ago) received it?s beginning from another source.
|
Article 2: Emergence - more than the sum of parts An important concept in aLife is emergent properties. "An emergent property is created when something becomes more than the sum of its parts." As Mr. Dean notes, life can be seen as "the most extreme example of an emergent property. . . .Half a human does not work without the other half, but as a whole we are capable of very complex behaviour. [sic]" Another example provided is the American phone system. Due to it?s complexity, functions which were not designed to be performed and operations which were unintended occurred and continue to occur. After researching these events, it was concluded that "they were the results of emergence." In terms of aLife, sets of instructions are provided to the system for functions to be performed when various factors are present (or absent). In this manner, behaviors for specific situational circumstances are preprogrammed. However, when numerous (or just more than a single) situational factors are combined, the resulting action(s) may have been unintended and/or unforeseen by the programmer. In this manner, the subjects in the programmed environment evolve, replicate, and adapt to their surroundings.
|
Article 3: Complexity and Entropy Mr. Dean states that "Life is about the fight against entropy." Various definitions for this term are provided in his article. To put it simply, entropy could be seen as reaching a static state in a closed system as energy is used and the amount of available energy decreases. Nature is not viewed as a closed system. "As other systems lose information to the surrounding environment (information can be seen as heat or other quantities) life has not only to keep hold of its information but also increase its amount of information. It does this by absorbing, or eating, energy from the surrounding environment." Thus to achieve entropy in a life system could signify death. In the words of Mr. Dean, "If entropy had its way the whole universe would be a soup of equal density, temperature and composition." Perhaps other life forms could survive if the earth were to achieve "equal density, temperature and composition", however, it would be extremely difficult, if not impossible, for the current human species. The significance of entropy for aLife is in the arena of survival. As aLife creatures evolve, replicate, and adapt to their environments, they must acquire resources from their surroundings. Skills in the form of instructions must be provided to allow them to attain the necessary nourishment. If aLife beings become unable to seek out and procure sustenance due to mutation or other means, then the entity will cease to exist. A complex system based on the research of Chris Langton, exists between a state which is chaotic and one which is frozen. A chaotic state is one in which stability is very difficult to achieve, and thus impractical for life. The frozen state being immobile, is also not feasible for life forms. An aLife system must be configured to be complex so that a relative stability is able to be maintained and life continues to evolve.
|
Article 4: Chaos Theory Chaos theory explains that "apparently random events can be represented by a simple computation which, when iterated, produce complex results." In other words, occurrences which appear to be random can be examined through the use of mathematics. In this process of analysis the "random" incidents are shown to follow logically (mathematically) from the preceding variables. As a very simplistic illustration, imagine a shiny red ball on the ground. The ball might not have been there a minute ago. And there might not be an individual in plain sight who could have caused the ball to appear at that spot. If however, the actual path of the ball could be traced, the variables which acted on that ball would explain why it surfaced at that point on the ground. The variables could be a) if it was thrown, in what direction and how fast, b) if it fell, from what height and angle, c) the wind conditions at the time, and d) particulars of the current environment (weather, hard or soft landing surface, etc.).
A chaotic system arises from the iteration of events and feedback. Feedback is a process whereby values produced from a stage are used in the computation of the next stage. An example of a chaotic system in nature is the sea which is driven "by a finite amount of inputs (wind, sun, tides, land masses, etc.)." As for an example which is the product of the human species, consider the financial world markets. "These, from the outside, appear to be random and unpredictable, but, as we know, each financial decision is a conscious decision and is far from random. Each step of the way computations are used, along with previous knowledge to make these decisions. Yet, due to the sheer complexity of dependencies, the results can often appear to be far from rational."
In the application of aLife, chaos theory is instrumental in reproducing behaviors seen in nature. "A chaotic system can only be predicted if all the inputs into the system are known and all the rules [of] the system are know [sic]. If only the rules are known you can reproduce the behaviour [sic] of something but not the exact occurrence that has been produced before." Thus, to create a system which will predict the motion of the waves, you must be aware of not only the various variables (such as wind speed, gravitational pull, etc.) acting on the waves, but also the manner in which these variables interact with the waves.
|
Article 5: Cellular Automata Based on research by John von Neumann, a cellular automata is conceived to be "an array of 'cells' that interact with their neighbors." The arrays can assume diverse dimensions, with each cell possessing a state of information which could be represented as a variable or property. The cell undergoes a change of state based on the instructions which were programmed for the cell and the environmental stimulus? from the adjoining cells. Depending on input from the surroundings and the computerized rules, the change of state which the cell experiences might include transmitting information to it?s adjacent cells, who in turn process the data and continue the "chain reaction". Mr. Dean compares the cellular automata functioning to natural cells. "For example in an embryo messages are spread by timed releases of chemicals by cells which tell a cell which type of cell it should be. . . . Different doses of chemicals and in different combinations cause a cell to act in different ways. Sometime [sic] small changes in chemical density can lead to radical changes in cell formation. This is how cell barriers are formed. . . .[and] complex forms, like a human baby, can be built up." In terms of aLife application, cellular automata assists in conceiving and explaining how information is passed between various components in a system. The ability to emulate transfer of data and processing of those details within the module to determine an outcome which may affect other segments allows programs to recreate patterns seen in the natural environment. Through regeneration of the characteristics, an aLife system may take on an energy form which could be described as being "alive".
|
Article 6: Creatures Creatures are the actual beings constructed to emulate aLife. Some creatures are designed with predestined behavior patterns and others are allowed more flexibility in their environmental interaction. Examples of aLife creatures given by Mr. Dean are as shown in the table below.
The aLife creations have been used in our daily lives, and in many applications we may not be aware of the impact of aLife designs. In movies where crowd formation and movement are shown, the animation of the group may have been designed using schemes such as Craig Reynold?s Boids. The logic behind the movement exhibited in the Boids environment has also assisted in the design of buildings. Allowing the designers to simulate how a group of individuals may move to escape a burning building would enable them to determine where exits should be located and how many will be required. In the aLife creations Neorotic, Polyworld, and the work of Karl Sims, these environments could be used to show how changes in the world?s temperatures, climates, etc. would impact on the survival of various species. ALife is still an emerging field. As potential problems are envisioned, aLife creations could be used as aids to determine the best course of action(s) to eliminate or minimize the predicaments.
|
Article 7: Neural Networks Artificial intelligence (AI) is the logical processing of creatures which allows the designed beings to function. A neural network is one approach that is used in mapping out the AI reasoning. The neural network could be compared to the operation of the brain. It is a collection of nodes called neurons which are interconnected, with each neuron possessing multiple input and output channels. The neuron passes information to its neighbors based on the deduction scheme implemented, information that it has gathered from previous interaction with it?s environment, and the surrounding input from adjacent neurons. The adjoining neurons in turn process the data and continue the "chain reaction". The aspect which I personally feel qualifies this method as "intelligence," is the ability to "learn" from it?s environment. Various AI programming languages allow values or combinations of values to be stored in the system?s "memory" that allows the program to recognize if it has encountered the current situation previously, and if so, what resulted from those circumstances. With this knowledge, the being combines feedback from it?s neighbors and it?s pre-programmed logic to determine how to react in the current situation. As Mr. Dean stated, "The [neural] net has to receive enough information through the input to be able to make the correct assumptions." If inaccurate or incomplete information is provided, then the resulting actions will not produce the desired outcome. A good illustration is provided by Mr. Dean in his article :
|
Article 8: Evolution and Evolutionary Computing Evolution is a process through which creatures adapt to their environment. Behaviors and characteristics of these beings which assist in survival and replication are reinforced and strengthened for future generations. The responses and attributes which are maladaptive to continued life and reproduction are reduced or eliminated in subsequent generations, or the particular species strain ceases to exist. An example of an aLife system mimicking evolution is the Biomorph program conceived by Richard Dawkins. The system uses mutation and manual selection to breed the Biomorph creatures. As stated by Mr. Dean:
|
Article 9: L-Systems - The systems of growth L-systems were conceived by Arstid Lindenmayer based on the schemes present in cellular automata. L-systems concentrate on "the emulation of natural growth" as can be seen in the expansion patterns of plants. The four elements that comprise an L-system are explained in the table below.
An example of a one-dimensional L-system provided in the article is as follows:
In each of the stages shown in the iteration table above, the variables can be given commands to be executed. As an illustration, each time the variable A is encountered (as in Stage 0), the system could draw a one-inch line to the right of the current position. When a variable B is detected (as in Stage 1), the computer could draw a one-inch line to the left of the current position. In terms of practical applications, the L-system logic could be used in home designing to determine what types of plants should be utilized, where they should be placed, and how many of each type to use. A simulation of the various growth patterns based on factors such as gravity, temperature, and climate could be taken into account to visualize how the environment would appear in a future time period.
|
Article 10: People - Visionaries and Evangelists Selected individuals that Mr. Dean considers as "visionaries and evangelists" in the field of aLife are briefly shown in the table below:
I was happily pleased with the advances that have been made in the fields of AI and aLife through the work of individuals such as those named in the above table. From my introduction to AI and aLife over a decade ago, I was fascinated with the innumerable possibilities that could be envisioned to apply these concepts in manners which would benefit our world. I personally feel that we have only scratched the surface of the capabilities of AI and aLife. As future obstacles are predicted, I hope that AI and aLife will be optimized to eliminate or minimize difficulties.
|
Article 11: The Uses and Abuses of aLife Three general categories of uses for aLife mentioned by Mr. Dean are as personal assistants, intelligent agents, and entertainment. In the arena of personal assistants, aLife could be programmed to filter the multitude of information that we receive so that what is actually viewed by each individual are the details that s/he requested. This would minimize the amount of spam (unsolicited mail) that is received. An example of intelligent agents that would be familiar to those taking Dr. James? courses would be search engines. The agents basically browse through the web pages and provide links to those pages which meet selected criteria. To illustrate, an agent could be assigned to locate all sites which contain information on aLife and will generate links to the pages which were discovered. The generated links will be displayed based on a hierarchy depending on how well each site met the pre-determined criteria. Concerning aLife and entertainment, interactive games are the most easily recognized example. As Mr. Dean accurately states, "Wether [sic] the character or creature you are interacting with is a friend or foe a game or piece of interaction is far more interesting if that character behaves in an intelligent way." Thus, instead of pre-determined interaction with your desired game, the system will be able to "learn" from your method of decision-making and thereby result in a session which is challenging to each individual in every encounter with that game. Although I felt that the uses explored for aLife were good examples, I was disappointed that other practical means of utilizing aLife were not mentioned. Applications such as describing and predicting chaotic systems in nature, designing buildings to enable inhabitants to safely escape hazards including fire, and the ability to simulate evolution experiments which would allow us to potentially avoid foreseen problems before they occur. Two general categories of aLife abuse explored by Mr. Dean include war and computer viruses. In the domain of war, military intelligence utilizes aLife to strengthen their fighting strategies. According to Mr. Dean, "It appears that the Defence Advanced Research Projects Agency of America has already got devices on its drawing board that takes the people out of combat or turns a single soldier into a battalion." The danger that I see in this application of aLife is the added desensitization which occurs when enemies are killed from afar. Although I don?t condone violence or war, if the "enemy" is right in front of you, you are more likely to realize that the "enemy" is another human being?someone capable of emotions, thoughts, being a part of a family. Removing the "human factor" from the issue makes it too easy to use unnecessary force because the actual impact of that force is unknown to the operator. I do realize the other side of the coin?that aLife may potentially reduce "our" casualties in the event of a war. However, "we" are not the only intelligent humans on this planet. If we possess the technology capable of mass destruction, we must also be aware that our "enemies" could have access to those same innovations. Perhaps our efforts in aLife war applications would be more suitable if we create programs which are designed to save lives instead of destroying them. As for computer viruses, there has been considerable debate as to whether they qualify as aLife. Personally I feel that computer viruses possess the capability to become aLife with AI. If you?re interested in an overview of computer viruses, check out my Report on the Psychology of Computer Viruses. If you?re just interested in my opinions on computer viruses as aLife, see the section Significance of Computer Viruses.
|
Article 12: About the aLife Interactive Essay Project Mr. Dean created the aLife Interactive Essay out of his passion for aLife. His ongoing maintenance includes a future conversion to style sheets, graphic additions, and interactive programs. As mentioned in my introduction, I feel that this site was configured in a style which would allow an individual to either acquire a fundamental level of knowledge on the entire scope of aLife, or just concentrate on a particular portion depending on the level of expertise that s/he possessed on this topic. I was extremely pleased with the conversational tone which Mr. Dean employed throughout his interactive essay, which explains the various concepts in easy-to-grasp wordings. My only criticism of this pages lies in a few misspelled words and minor grammatical oversights which probably resulted from numerous revisions. However, these points are insignificant compared to the vast amount of knowledge contained in this interactive essay.
|
How others in this generation compiled their findings
Overall, I feel that Generation 8 has done a terrific job on their second reports. Although no one else has completed a report on aLife or AI, I did feel that the following reports were rather outstanding in their own rights: Holly Ishikawa made a good presentation and her use of colors was excellently chosen. Alona Lei Tabios designed a report in which the layout was effective and easy to comprehend, all without graphics. For a look at great reports on computer viruses, check here.
|
Conclusions In studying the psychology of artificial life (aLife) and artificial intelligence (AI), I would suggest that future generations begin by reading this report and the actual aLife Interactive Guide. After gaining a general idea of the concepts involved in aLife and AI, I would advocate studying the links contained in the resources section. To actually take this report to the "next step", I would investigate the current operations of various aLife and AI labs to determine the directions and focus of recent research. Another direction could be to discover the world of computer viruses, and explore the current merging of this area to AI and aLife. For an introduction to computer viruses, you can start with my report on this area.
|
