Class project of SEIS 610 Software Engineering at the University of St. Thomas.
For a description of what Genetic Programming is see the wikipedia article here: http://en.wikipedia.org/wiki/Genetic_programming.
Given a set of test data, where f(x) = (x^2 - 1) / 2, randomly generate a set of solutions. Evaluate each solution to see how closely it matches f(x) = (x^2 - 1) / 2. The genetic program should evolve the solutions until one is found that is equivalent to f(x) = (x^2 - 1) / 2.
A genetic program in C++ using binary trees.
Training data used: (-16, 128.5) (-8, 32.5) (-4, 8.5) (-2, 2.5) (0, 0.5) (1, 1) (3, 5) (5, 13) (9, 41) (17, 145)