Genetic Programming in Mathematica
Genetic Programming is generally considered to be a branch of Artificial
Intelligence. it is a generalisation of Genetic Algorithms. together with
other methods, these fall into the sub-category called Automatic
the basic theory is that it should be possible to have programs create other
programs. this would mean that some tasks would not have to coded manually.
instead a computer could be "asked" to generate a solution to a problem in
the form of a computer program.
in Genetic Programming, the solution is evolved by applying the principles
of evolution and Darwinian survival of the fittest. a group of random
solutions is first created and these are then changed by randomly breaking
up and exchanging portions of the solutions. the degree to which a solution
solves a problem correctly determines that solution's ability to interact
with other solutions. eventually the group of solutions produces one that is
i completed my Masters thesis in 1997.
hussein, Miloslav Hajek
Suleman, H., and M. Hajek (1996).
Parallelism: An Effective Genetic Programming Implementation on Low-powered Mathematica Workstations,
in Proceedings of The 1996 National Research and Development Conference: SAICSIT 1996,
26-27 September, University of Natal, Durban.
Suleman, H. (1997).
Genetic Programming in Mathematica,
M.Sc. thesis, University of Durban-Westville.