Size fair and homologous tree crossovers
WilliamB Langdon · Centrum Wiskunde & Informatica (CWI), the national research institute for mathematics and computer science in the Netherlands · 1999
Size fair and homologous crossover genetic operators for tree based genetic programming are described and tested. Both produce considerably reduced increases in program size (i.e. less bloat) and no detrimental effect on GP performance. GP search spaces are partitioned by the ridge in the number of program v. their size and depth. While search efficiency is little effected by initial conditions, these do strongly influence which half of the search space is searched. However a ramped uniform random initialisation is described which straddles the ridge. With subtree crossover trees increase about one level per generation leading to sub-quadratic bloat in program length. 1991 Mathematics Subject Classification: 68N05, 68Q25, 68R10, 68T05, 68T20 1991 Computing Reviews Classification System: F.2.m, F.3.m, G.1.6, G.2.1, G.2.2, G.3, I.2.2, I.2.6, I.2.8 Keywords and Phrases: genetic algorithms, genetic programming, bloat reduction, evolution of shape, subquadratic length growth, linear dept...