Backward chaining evolutionary algorithms (BC-EA)offer the prospect of runtime efficiency savings byreducing the number of fitness evaluations withoutsignificantly changing the course of genetic algorithm(GA) or genetic programming (GP) runs.poli05:_tourn_selec_iterat_coupon_probl described howBC-EA does this by avoiding the generation andevaluation of individuals which never appear inselection tournaments. (Poli,2005) suggested thelargest savings occur in very large populations, shortruns and small tournament sizes. It gave some evidenceof the actual savings in fixed-length binary GAs. Herewe provide a generational GP implementation of BC-EAand empirically investigate its efficiency, in terms ofboth fitness evaluations and effectiveness, withmutation and two offspring crossover.
CITATION STYLE
Poli, R., & Langdon, W. B. (2006). Running Genetic Programming Backwards. In Genetic Programming Theory and Practice III (pp. 125–140). Kluwer Academic Publishers. https://doi.org/10.1007/0-387-28111-8_9
Mendeley helps you to discover research relevant for your work.