Handling deceptive problems using a different genetic search

In recent years, several studies have been devoted to the design of problems with different degrees of deception in order to investigate the performance of GAs. This paper presents a different genetic approach, called the structured genetic algorithm (sGA) for solving GA-deceptive problems. The structured GA uses an hierarchical encoding and a gene expression mechanism in its overspecified chromosomal representation. The paper reported some experimental results which demonstrated that on using a different chromosomal representation (as in sGA), the genetic search becomes more robust and can easily handle so-called GA-deceptive problems