论文信息 - Boolean Functions Fitness Spaces

Boolean Functions Fitness Spaces

We investigate the distribution of performance of the Boolean functions of 3 Boolean inputs (particularly that of the parity functions), the always-on-6 and even-6 parity functions. We use enumeration, uniform Monte-Carlo random sampling and sampling random full trees. As expected XOR dramatically changes the fitness distributions. In all cases once some minimum size threshold has been exceeded, the distribution of performance is approximately independent of program size. However the distribution of the performance of full trees is different from that of asymmetric trees and varies with tree depth.

Riccardo Poli | William B. Langdon | W. Langdon | R. Poli

[1] L. Alonso,et al. Random Generation of Trees , 1995, Springer US.

[2] Riccardo Poli,et al. On the Search Properties of Different Crossover Operators in Genetic Programming , 2001 .

[3] J. K. Kinnear,et al. Advances in Genetic Programming , 1994 .

[4] Justinian Rosca,et al. Hierarchical learning with procedural abstraction mechanisms , 1997 .

[5] Riccardo Poli,et al. The evolution of size and shape , 1999 .

[6] Kevin J. Lang. Hill Climbing Beats Genetic Search on a Boolean Circuit Synthesis Problem of Koza's , 1995, ICML.

[7] David H. Wolpert,et al. No free lunch theorems for optimization , 1997, IEEE Trans. Evol. Comput..

[8] Riccardo Poli,et al. Why “Building Blocks” Don’t Work on Parity Problems , 1998 .

[9] William B. Langdon,et al. Scaling of Program Fitness Spaces , 1999, Evolutionary Computation.

[10] Hitoshi Iba,et al. Genetic Programming 1998: Proceedings of the Third Annual Conference , 1999, IEEE Trans. Evol. Comput..

[11] Riccardo Poli,et al. Why Ants are Hard , 1998 .

[12] John R. Koza,et al. Genetic programming - on the programming of computers by means of natural selection , 1993, Complex adaptive systems.

[13] Riccardo Poli,et al. Fitness Causes Bloat , 1998 .