Avoiding Two-Bit Crossovers in Genetic Programming
Abstract:
We investigate the utility of weighting the crossover points in genetic programming. The depth-fair crossover (DFC) operator is introduced as an alternative to the standard 90=10 weight heuristic. The DFC weight heuristic performs better that the standard 90=10 weight heuristic in the clique domain. Preliminary results also indicate it will perform better in other applications. 1 Introduction Genetic programming (GP) can be used to solve a variety of NP-hard and NP-complete problems. One such problem is that of clique detection (Haynes 1996, Soule et al. 1996). When detecting both the max clique and the clique cover, the results obtained by a standard genetic algorithm (GA) are better than the results obtained by a standard genetic programming algorithm (Haynes 1998). A two-point crossover in a GA picks any two points on the chromosome to randomly crossover and each point is equally likely to be chosen for crossover. In GP this is not the case. The standard chromosome of GP is a par...
Citations
| 7271 | Computers and Intractability - A Guide to the Theory of NP-Completeness – Garey, Johnson - 1979 |
| 5180 | Genetic Algorithms – Goldberg - 1989 |
| 172 | Genetic Programming: On the Programming of Computers by Natural Selection – Koza - 1992 |
| 139 | Adapting operator probabilities in genetic algorithms – Davis - 1989 |
| 49 | Two self-adaptive crossover operations for genetic programming – Angeline - 1996 |
| 22 | Duplication of coding segments in genetic programming – Haynes - 1996 |
| 7 | Two ways of discovering the size and shape of a computer program to solve a problem – Koza - 1995 |
| 4 | Dynamics of Genetic Programming and Chaotic Time Series Prediction – Mulloy, Riolo, et al. - 1996 |
| 3 | Collective Adaptation: The Sharing of Building Blocks – Haynes - 1998 |
| 2 | Causality in genetic programming. In: Genetic Algorithms – Rosca, Ballard - 1995 |

