Duplicate genotypes in a genetic algorithm

Abstract

In a simple genetic algorithm (GA), building-block accumulation stops when the population converges. Premature convergence is a clear indication that this accumulation process has occurred insufficiently. In such a case, the final population members will not contain the best combination of building blocks. It is common practice to remove duplicate-population genotypes in a steady-state GA. This allows the process of building-block accumulation to continue for a greater period of time compared with a regular GA and typically results in better final population solutions. A striking observation is the fact that duplicates degrade GA quality early in a GA evolution run. The results show that diversity loss through duplicates is a serious weakness in the steady-state GA model.