Information Technology Reference
In-Depth Information
genotype is just the bit strings, the parameters of which are exactly like the
weight values or thresholds in the artificial neural network.
4. Robots
The robots offer the approaches totally different from life-like systems. With
respect to the form they are similar to living beings, but they have no
reproduction capability. They can predict the future through deducing. The robots
have a lot of properties related to life—complexity, organism-integration,
stimulation, and movablity.
5. Autocatalytic network
In contrast to high-technology robots, the autocatalytic network is designed and
constructed by people, which produces simple life-like systems with the
reproducing (or self-evolving) capabilities in the simulating chemical test tube.
This kind of systems have two active research phrases—autocatalytic network
and cellular automata.
Autocatalytic network has some transformation, the most complete one of
which is the theory of hypercycle. It is linked to the response component (RNA)
network, which allows the consistent function coupled evolution, and
self-replicating entity. Herein, the hypercycle network is taken as an example to
describe all the natural life systems. Eigen has already developed the theory of
deducing on hypercycle network, which can increase its complexity
spontaneously, and know when collapses. Its mathematics processing is based on
continuous reaction kinematics. Suppose there is no structure of space in the
complete mixing of the element, it is superior to the model of cellular automata.
6. Cellular automata
It is a cell array, each cell of which is in discrete structure. According to the rules
given in advance, the states can change with time. The transfer rules are to
compute the present state of a cell and states of its neighbors by the array, all
cells are updated spontaneously. Both of autocatalytic networks and cellular
automata produce the modes of self-organization and self-replication, which do
not change with time, but the mathematical modes of processing them are
different.
The history of cellular automata is brilliant. Cellular automata are invented by
von Neumann in 1940, which can be regarded as not only an important method
of understanding the natural systems (natural automata) in the forms of
