Artificial Life

Artificial Life

The phrase Artificial Life or alife is coined by Christopher Langton in the Proceedings of the Interdisciplinary Workshop on the Synthesis and Simulation of Living Systems [1]. The field owes its deepest intellectual roots to John von Neumann and Norbert Wiener [2]. Von Neumann [3] was the first to design an Artificial Life model without referring to it as such. He attempted to understand the fundamental properties of living systems by creating his self-reproducing Cellular Automata (CA) that exhibited those properties. Wiener [4] at about the same time started applying information theory and the analysis of self-regulatory processes (homeostasis) to the study of living systems. According to Bedau [2], both the constructive and abstract methodology of cellular automata and the abstract and material-independent methodology of information theory still exemplify much of alife.

Artificial Life as defined by Langton [1], is “…the study of man-made systems that exhibit behaviours characteristic of natural living systems. It complements the traditional biological sciences concerned with the analysis of living organisms by attempting to synthesize life-like behaviours within computers and other artificial media. By extending the empirical foundation upon which biology is based beyond the carbon-chain life that has evolved on Earth, Artificial Life can contribute to theoretical biology by locating life-as-we-know-it within the larger picture of life-as-it-could-be.” It is simply “life made by humans rather than by nature”.  However, in later years, Langton states that “Artificial Life is many things to many people” [5]; he also suggests that “It is probably safe to say that the field as a whole represents attempt to increase vastly the role of synthesis in the study of biological phenomena.” Rietman [6], on ‘Creating Artificial Life’ gave a short list of biological life that might consist of reproduction, metabolism, evolution, growth, self-repair, and adaptability. A practical goal of alife is to find a mechanism for an evolutionary process to be used in the automatic design and creation of artefact [7].

The history of the formal studies in AI is significantly earlier than alife. As such some have attempted to classify alife research as being a subfield of AI. Even though studies in alife overlap with those in AI, both approach the subject rather differently. The studies in AI aim to model the human mind whereas alife attempts at modelling life. The former uses the top-down approach whereas the latter uses the bottom-up approach of emergent systems. In fact, research in Complex Adaptive Systems (CAS) and alife are parallel.

Artificial Life research is divided into two main groups. The Strong alife position suggests that “life is a process which can be abstracted away from any particular medium” (John von Neumann) and can be created by the use of inorganic matter via some simple initial conditions or simple mechanical rules. The Weak alife position suggests that life cannot be generated outside of a carbon-based chemical solution but its processes could be understood by mimicking it in computer simulations.

Bedau [8] identified the three branches of alife:

  • Soft alife creates simulations or other purely digital constructions that exhibit life-like behaviour
  • Hard alife produces hardware implementations of life-like systems
  • Wet alife synthesises living systems out of biochemical substances

The work in alife addresses two issues:

  • The study of life beyond the carbon-chain chemistry in biological life
  • The application of the principles of life for problem solving

Biology is the scientific study of carbon-based life forms. The fundamental obstacle in theoretical biology is that it is impossible to derive general principles from single examples. In order to derive general theories and to distinguish the essential properties of life, comparisons had to be made from many instances of life. Time is also an obstacle. The study of the life-cycles of organisms and their genetic descent requires the element of time. The other factor is data-noise. The compounds of data that can be gathered in the study of life inevitably results in information noises. This obstacle frustrates analysis of matter that requires unpolluted datasets. Artificial Life resolves these issues by creating alternative life-forms within computers. Computer simulation has permitted a new approach to the study of evolution and natural systems. According to Mitchell [9], simulation can be controlled, repeated to see how the modification of certain parameters changes the behaviour of the simulation, and run for many simulated generations. Such great control over synthesised life within computers has given researchers the means to eliminate the limitations of time. It has also allowed the omission of information noises by filtering parameters that are not required so that a better understanding of life can be realised as a result of an unpolluted computer generated environment.

According to Christopher Langton, Artificial Life is “the study of synthetic systems that exhibit behaviors characteristic of natural living systems. It complements the traditional biological sciences concerned with the analysis of living organisms by attempting to synthesise life-like behavior within computers and other artificial media. By extending the empirical foundation upon which biology is based beyond the carbon-chain life that evolved on Earth, Artificial Life can contribute to theoretical biology by locating life-as-we-know-it within the larger picture of life-as-it-could-be” [10]. Studies in alife are similar to biology – alife is an experimental science. It attempts to understand the mechanisms behind living systems by synthesising them so that the structure and function of these systems may be understood and applied. Langton again states that “By extending the horizons of empirical research in biology beyond the territory currently circumscribed by life-as-we-know-it, the study of Artificial Life gives us access to the domain of life-as-it-could-be, and it is within this vastly larger domain that we must ground general theories of biology and in which we will discover novel and practical applications of biology in our engineering endeavors.”

Artificial Life is concerned with generating life-like behaviours within computers. Since the study focuses on the behaviours of life, it involves identifying the intrinsic mechanisms that generate such behaviours. These mechanisms are often very simple rules within an organism out of which complex behaviour emerges. Or on a higher level, the mechanism itself may be the simple behaviours of an organism. Together as cooperative entities, these organisms worked together as a larger ‘organism’ for the survival of the colony, yet from very basic pre-programmed rules in the monotony or variety of individual behaviours. The phenomena or intelligence that emerges as a result of the simple interaction between individual entities is called emergence [11, 12], a central concept supporting studies in alife.

Systems that exhibit emergent behaviours are commonly expressed in the sentence “the whole is greater than the sum of its parts” [13]. The term emergence was first used by an English philosopher G.H. Lewes [14] over a hundred years ago and has been subsequently studied in philosophy [15-18]. However, the process that issued in emergent behaviour or property was unknown to them until the advent of computers and the initiation of the theory of complexity [9-11] coupled with experiments on Cellular Automata (CA) [13, 22-25]. Studies in CA showed that simple rules in computer agents could give rise to complex behaviours. A particular CA ‘Life rule’ [16] invented by John Conway in the 1970s called the Game of Life demonstrated the idea of emergence where simple interactions between entities resulted in objects and patterns that are surprising and counter-intuitive. Referring to the definition of the term ‘emergence’, John Holland [11], a pioneer in complex systems and nonlinear science advised that “It is unlikely that a topic as complicated as emergence will submit meekly to a concise definition, and I have no such definition to offer”, however, he adds that “The hallmark of emergence is this sense of much coming from little… where the behaviour of the whole is much more complex than the behaviour of the parts.” Another definition [27] states that emergence is “…the process by which patterns or global-level structures arise from interactive local-level processes. This “structure” or “pattern” cannot be understood or predicted from the behaviour or properties of the component units alone.” In a more elaborate sentence given by Stacey [18], “Emergence is the production of global patterns of behaviour by agents in a complex system interacting according to their own local rules of behaviour, without intending the global patterns of behaviour that come about. In emergence, global patterns cannot be predicted from the local rules of behaviour that produce them. To put it another way, global patterns cannot be reduced to individual behaviour.”

The success of alife in problem solving is identified by the possibility of emulating nature’s properties of decentralisation, self-organisation, self-assembly, self-producing, and self-reproduction, all of which are key concepts in the science of alife. A decentralised system [29] relies on lateral relationships for decision making instead of on the hierarchical structure of command or force. Self-organisation [13, 30] refers to systems that manage itself and increases in productivity automatically without guidance from an external source. In systems with characteristics of self-assembly, patterns are seen to form from simple disordered components. Self-producing or autopoiesis [31] connotes the idea that certain types of system continuously produce their constituents, their own components, which then participate in these same production processes. An autopoietic system has a circular organisation, which closes on itself, its outputs becoming its own inputs. Such systems possess a degree of autonomy from their environment since their own operations ensure, within limits, their future continuation. In Self-replication [23, 32, 33], entities makes copies of themselves.

Resnick [29] presents evidence of a trend of decentralisation across important domains in nature, organisation, society, science, and philosophy. He states that orderly patterns can arise from simple, local interactions and that many things are “organised without an organisation, coordinated without a coordinator”. In the same way, Farmer and Packard [34] noted that in self-organising systems, patterns emerge out of lower-level randomness. Furthermore, Charles Darwin [35] asserted that order and complexity arises from decentralised processes of variation and selection. Langton [36], the founder of alife, stated that in order to model complex systems like life, the most promising approach is to dispense “with the notion of a centralised global controller” and to focus “on mechanisms for the distributed control of behaviour”.

The characteristics inherent in many decentralised systems are nonlinear, and the nonlinearity, as characterised by natural and alife systems is what makes the outcome more than the sum of their parts – forming dynamic patterns, collective intelligence and unpredictable behaviours from the lateral interactions between similar entities. This is the difference between the bottom-up approach in alife and the top-down methods in AI and many older man-made systems. By employing the bottom-up approach observed in nature, not only can we discover how certain systems work, but many complex problems previously impenetrable with top-down methods can now be readily solved.

Eugene Ch’ng, June 2005

References:

[1]    C.G. Langton, ed. Artificial Life, Proceedings of an Interdisciplinary Workshop on the Synthesis and Simulation of Living Systems. 1989, Addison-Wesley Publishing: Redwood City.

[2]    M.A. Bedau, Artificial Life: Organization, adaptation and complexity from the bottom up.. Trends in Cognitive Sciences, 2003. 7(11): p. 505-512.

[3]    J. von Neumann, Burks, A.W., Theory of Self-Reproducing Automata. (Urbana IL: University of Illinois Press, 1966).

[4]    N. Wiener, Cybernetics, or Control and Communication i nthe Animal and the Machine. Wiley, 1948).

[5]    C.G. Langton, ed. Artificial Life: An Overview. 1995, MIT Press: Cambridge.

[6]    E. Rietman, Creating Artificial Life: Self Organization. (Blue Ridge Summit, PA: Windcrest Books/McGraw Hill, 1993).

[7]    J. Vaario, Artificial life primer (Technical Report TR-H-033). (Kyoto: ATR Human Information Processing Research Labs, 1993).

[8]    M.A. Bedau, Artificial Life: Organization, adaptation and complexity from the bottom up.. Trends in Cognitive Sciences, 2003. 7(11): p. 505-512.

[9]    M. Mitchell, An Introduction to Genetic Algorithms (Complex Adaptive Systems). The MIT Press, 1998).

[10]    C.G. Langton, Artificial Life. (Boston, MA: Addison-Wesley Longman Publishing Co., Inc., 1990).

[11]    J. H. Holland, Emergence from Chaos to order. (Oxford: Oxford University Press, 1998).

[12]    S Johnson, Emergence: The Connected Lives of Ants, Brains, Cities, and Software. Sribner, 2002).

[13]    S. Kauffman, At Home in the Universe: The search for laws of complexity. (Harmondsworth: Penguin, 1996).

[14]    G.H. Lewes, Problems of Life and Mind. (London: Kegan Paul, Trench, Turbner, & Co., 1875).

[15]    B.P. McLaughlin, The Rise and Fall of British Emergentism. Emergence or Reduction?: Essays on the Prospects of Nonreductive Physicalism. (Berlin: Walter de Gruyter, 1992).

[16]    J.S. Mill, System of Logic. 8th ed (London: Longmans, Green, Reader, and Dyer, 1843).

[17]    E. Nagel, The Structure of Science. (New York: Harcourt, Brace and Wilson, 1961).

[18]    J. Goldstein, Emergence as a Construct: History and Issues. 1, 1999. 1: p. 49-72.

[19]    M. M. Waldrop, Complexity: The Emerging Science at the Edge of Order and Chaos. (London: Viking, 1993).

[20]    R. Lewin, Complexity: Life on the Edge of Chaos. (London: Phoenix, 1993).

[21]    J. H. Holland, Hidden Order: How adaptation builds complexity. Helix Books, Addison-Wesley Publishing, 1995).

[22]    J. von Neumann, Burks, A.W., Theory of Self-Reproducing Automata. (Urbana IL: University of Illinois Press, 1966).

[23]    C.G. Langton, “Studying Artificial Life with Cellular Automata, Evolution, Games and Learning: Models of Adaptation in Machines and Nature, Proceedings of the Fifth Annual Conference of the Centre for Nonlinear Studies. 1986. Los Alamos: Amsterdam: North-Holland,p. 120-149

[24]    T. Toffoli and N. Margolus, Cellular automata machines: A new environment for modelling. (Cambridge: MIT Press, 1987).

[25]    H. Gutowitz, Cellular automata: Theory and experiment. Physica D, 1991. 45: p. 1-3.

[26]    M. Resnick, Silverman, B., The Facts of Life. 1996, http://llk.media.mit.edu/projects/emergence/life-intro.html Last accessed 3 January 2006, 2006.

[27]    K. Mihata, The Persistence of ‘Emergence’, in Chaos, Complexity & Sociology: Myths, Models & Theories, A.E. Raymond, Horsfall, S., Lee., M.E., Editor. 1997, Thousand Oaks: California: Sage. p. 30-38.

[28]    R. Stacey, Complexity and Creativity in Organizations. (San Francisco: Berrett-Koehler, 1996).

[29]    M. Resnick, Turtles, Termites, and Traffic Jams: Explorations in Massively Parallel Microworlds. (Cambridge, Massachusetts: MIT Press, 1994).

[30]    S.A. Kauffman, The Origins of Order: Self-Organization and Selection in Evolution. (Oxford: Oxford University Press, 1993).

[31]    J. Mingers, Self-Producing Systems: Implications and Applications of Autopoiesis. (New York and London: Plenum Press, 1995).

[32]    M. Sipper, Fifty years of research on self-replication: an overview. Artificial Life, 1998. 4: p. 237-257.

[33]    M. Sipper and J.A. Reggia, Go forth and replicate. Sci. Am., 2001. 285: p. 34-43.

[34]    D. Farmer, Packard, N., Evolution, Games, and Learning: Models for Adaptations in Machines and Nature. Physica D, 1986. 22D(1).

[35]    C Darwin, The oigin of species by means of natural selection. (LONDON: JOHN MURRAY, 1859).

[36]    C.G. Langton, ed. Artificial Life, Proceedings of an Interdisciplinary Workshop on the Synthesis and Simulation of Living Systems. 1989, Addison-Wesley Publishing: Redwood City.

Leave a Reply

Your email address will not be published. Required fields are marked *