Sieves for Selection of Genes According to Desired Behaviors - Systematic Design for Emergence in Cellular Nonlinear Networks - page 99

knowing that some initial states may exists to contradict it. In terms of fuzzy logic
the Game of Life may be characterized as: “ A1 : Implosive exponent of growth”
AND “ A2 : Low value of the variance Var ” AND “ A3 : Large transient”, AND “ A4 :
Clustering coefficient close to 1”. An interesting question that arise is: Find CA
genes leading to similar genes. The equivalent “A” statement is: “CA behaving
like the Game of Life”.
Let us define the following sieve: , where: P x P x P x P x

A A 1 A 3 A 4

> @

, , and

P

exp 1 U 0 03

P

0 . 0 . tanh 10 Tr 0 .

1 3

,

P

exp 0 Clus 0 975

4

Such sieves were defined ignoring the Var parameter. A list of ID obtained for
an acceptable truth degree

T includes the “Game of Life” as well as other
13 genes: 6152, 11127, 11175, 11895, 13719, 14039, 15271, 23847, 44223,
97951, 238292, 239451, 245913, 250018.
As seen in Fig. 6.4 what is specific for most of these genes is a slow evolution
towards a globally stable or low periodic (ordered) state during which some short-
life compact formations (gliders) emerge. The size and the frequency of the
emerging gliders as well as the speed of convergence towards a global ordered
state depend on the particular gene and can be studied individually.

0 99

Fig. 6.4. The dynamics of a CA (ID = 11127) selected by the “Similar to Game of Life”
sieve. Note the emergence of gliders with a higher frequency during the first iterations. After
several thousands of iterations the CA evolves to a more ordered global state