June 26 - 30, 2004 Saturday to Wednesday Seattle, Washington, USA
MSA - Military and Security Applications of Evolutionary Computation
Parametric Regression Through Genetic Programming
Edwin Roger Banks James C. Hayes
Parametric regression in genetic programming can substantially speed up the search for solutions. In this paper parametric regression is applied to a minimum-time-to-target problem. The solution is equivalent to the classical brachistochrone. Two formulations were tried: a parametric regression and the classical symbolic regression formulation. The parametric approach was superior under a wide variety of conditions. We speculate the parametric approach is more generally applicable to other problems and suggest areas for more research.