Gene Deletion/Insertion

  GEP Book

  Home
  News
  Author
  Q&A
  Tutorials
  Downloads
  GEP Biblio
  Contacts

  Visit Gepsoft

C. FERREIRA

In J. M. Santos and A. Zapico, eds., Proceedings of the Argentine Symposium on Artificial Intelligence, pages 160-174, Santa Fe, Argentina, 2002.

Combinatorial Optimization by Gene Expression Programming: Inversion Revisited

The gene deletion/insertion operator is the second in importance of the analyzed combinatorial-specific operators (see Figure 1). This operator corresponds to the “insertion mutation” operator in the classification proposed by Larrañaga et al. (1998). Here, the designation “gene deletion/insertion” was chosen for three reasons: 1) to reflect the fact that the inserted element is previously deleted; 2) to emphasize that only one gene is deleted/inserted at a time; and 3) to distinguish this operator from the closely related operator “sequence deletion/insertion” described in section 3.4.

The gene deletion/insertion operator randomly selects the chromosome, the multigene family to be modified, the gene to transpose, and the insertion site. Each chromosome can only be modified once by this operator.

Consider the chromosome below composed of two multigene families, each with 13 members:

01234567890120123456789012
ifghabdecjklmklhcigdfejmba	(3.5)

Suppose gene 3 in MGF₁ was selected to transpose to site 7 (between genes 6 and 7). Then gene 3 is deleted in the place of origin and inserted between genes “d” and “e”, obtaining:

01234567890120123456789012
ifgabdhecjklmklhcigdfejmba	(3.6)

The deletion/insertion of genes when combined with more powerful operators such as inversion, might be useful to make finer adjustments. However, for all the problems analyzed in this work, the performance was higher if inversion worked alone.

Home | Contents | Previous | Next