NewEnvXOR {xega} | R Documentation |
Generate the problem environment EnvXOR
Description
NewEnvXOR()
generates the problem environment
for the XOR-Problem.
The problem environment provides an abstract interface
to the simple genetic programming algorithm.
ProblemEnv$f(parm)
defines the function we want to optimize.
A problem environment is a function factory with the following elements:
-
name()
: A string with the name of the environment. -
ProblemEnv$f(word)
: Function with theword
a word of the language (as a text string).
Should be provided by the user as a standard R-file.
Usage
NewEnvXOR()
Value
The problem environment:
-
$name
: The name of the problem environment. -
$f
: The fitness function. For this environment, fitness is defined as the number of correct test cases (correct function) and the inverse of the number of terminal symbols. The second part means that a boolean function with a fewer number of variables and logical functions is fitter than one with more variables and logical functions if both solve the same number of test cases.
See Also
Other Problem Environment:
Parabola2D
,
Parabola2DEarly
,
lau15
Examples
EnvXOR<-NewEnvXOR()
EnvXOR$name()
a2<-"OR(OR(D1, D2), (AND(NOT(D1), NOT(D2))))"
a3<-"OR(OR(D1, D2), AND(D1, D2))"
a4<-"AND(OR(D1,D2),NOT(AND(D1,D2)))"
gp4<-"(AND(AND(OR(D2,D1),NOT(AND(D1,D2))),(OR(D2,D1))))"
EnvXOR$f(a2)
EnvXOR$f(a3)
EnvXOR$f(a4)
EnvXOR$f(gp4)