A

Artificial Life

Artificial life (also known as "ALife") is an interdisciplinary study of life and lifelike processes by means of computer simulation and other methods. The goals of this activity include understanding and creating life and lifelike systems, and developing practical devices inspired by living systems. The study of artificial life aims to understand how life arises from non-life, to determine the potentials and limits of living systems, and to explain how life is connected to mind, machines, and culture.

The American computer scientist Christopher Langton coined the phrase "artificial life" in 1987, when he organized the first scientific conference explicitly devoted to this field. Before there were artificial life conferences, the simulation and synthesis of lifelike systems occurred in isolated pockets scattered across a variety of disciplines. The Hungarian-born physicist and mathematician John von Neumann (1903–1957) created the first artificial life model (without referring to it as such) in the 1940s. He produced a self-reproducing, computation-universal entity using cellular automata. Von Neumann was pursuing many of the questions that still drive artificial life today, such as understanding the spontaneous generation and evolution of complex adaptive structures.

Rather than modeling some existing living system, artificial life models are often intended to generate wholly new—and typically extremely simple—instances of lifelike phenomena. The simplest example of such a system is the so-called Game of Life devised by the British mathematician John Conway in the 1960s before the field of artificial life was conceived. Conway was trying to create a simple system that could generate complex self-organized structures.

The Game of Life is a two-state, two-dimensional cellular automaton. It takes place on a rectangular grid of cells, similar to a huge checkerboard. Time advances step by step. A cell's state at a given time is determined by the states of its eight neighboring cells according to the following simple "birth-death" rule: a "dead" cell becomes "alive" if and only if exactly three neighbors were just "alive," and a "living" cell "dies" if and only if fewer than two, or more than three, neighbors were just "alive." When all of the cells in the system are simultaneously updated again and again, a rich variety of complicated behavior is created and a complex zoo of dynamic structures can be identified and classified (blinkers, gliders, glider guns, logic switching circuits, etc.). It is even possible to construct a universal Turing machine in the Game of Life, by cunningly arranging the initial configuration of living cells. In such constructions, gliders perform a role of passing signals. Analyzing the computational potential of cellular automata on the basis of glider interactions has become a major direction in research. Like living systems, Conway's Game of Life exhibits a vivid hierarchy of dynamical self-organized structures. Its self-organization is not a representation of processes in the real world, but a wholly novel instance of this phenomenon.

To understand the interesting properties of living systems, von Neumann and Conway each used a constructive method. They created simple and abstract models that exhibited the kind of behavior they wanted to understand. Contemporary artificial life employs the same constructive methodology, often through the creation of computer models of living systems. This computer methodology has several virtues. Expressing a model in computer code requires precision and clarity, and it ensures that the mechanisms invoked in the model are feasible.

Artificial life is similar to artificial intelligence (AI). Both fields study natural phenomena through computational models, and most naturally occurring intelligent systems are, in fact, alive. Despite these similarities, AI and artificial life typically employ different modeling strategies. In most traditional artificial intelligence systems, events occur one by one (serially). A complicated, centralized controller typically makes decisions based on global information about all aspects of the system, and the controller's decisions have the potential to affect directly any aspect of the whole system.

This centralized, top-down architecture is quite unlike the structure of many natural living systems that exhibit complex autonomous behavior. Such systems are often parallel, distributed networks of relatively simple low-level "agents," and they all simultaneously interact with each other. Each agent's decisions are based on information about only its own local situation, which they affect.

In similar fashion, artificial life characteristically constructs massively parallel, bottom-up-specified systems of simple local agents. One repeats the simultaneous low-level interactions among the agents, and then observes what aggregate behavior emerges. These are sometimes called "agent-based" or "individual-based" models, because the system's global behavior arises out of the local interactions among a large collection of "agents" or "individuals." This kind of bottom-up architecture with a population of autonomous agents that follow simple local rules is also characteristic of the connectionist (parallel, distributed processing, neural networks) movement that swept through AI and cognitive science in the 1980s. In fact, the agents in many artificial life models are themselves controlled internally by simple neural networks.

Computer simulation in artificial life plays the role that observation and experiment play in more conventional science. The complex self-organizing behavior of Conway's Game of Life would never have been discovered without computer simulations of thousands of generations for millions of sites. Simulation of large-scale complex systems is the single most crucial development that has enabled the field of artificial life to flourish.

Living systems exhibit a variety of useful properties such as robustness, flexibility, and automatic adaptability. Some artificial life research aims to go beyond mere simulation by constructing novel physical devices that exhibit and exploit lifelike properties. Some of this engineering activity also has a theoretical motivation on the grounds that a full appreciation of life's distinctive properties can come only by creating and studying real physical devices. This engineering activity includes the construction of evolving hardware, in which biologically-inspired adaptive processes control the configuration of micro-electronic circuitry. Another example is biologically inspired robots, such as those robotic controllers automatically designed by evolutionary algorithms.