 |
    |
What Makes a Problem GP-Hard? Validating a Hypothesis of Structural Causes
Jason M. Daida, Hsiaolei Li, Ricky Tang, and Adam M. Hilss
Center for the Study of Complex Systems
and Space Physics Research Laboratory
The University of Michigan
2455 Hayward Avenue
Ann Arbor, Michigan 48109-2143
Abstract. This paper provides an empirical test of a hypothesis, which describes the effects of structural mechanisms in genetic programming. In doing so, the paper offers a test problem anticipated by this hypothesis. The problem is tunably difficult, but has this property because tuning is accomplished through changes in structure. Content is not involved in tuning. The results support a prediction of the hypothesis - that GP search space is significantly constrained as an outcome of structural mechanisms.
LNCS 2724, p. 1665 ff.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2003