Jumping Genes-mutators Can Rise Efficacy Of Evolutionary Search

Genetic Algorithms (GA) and Genetic Programming were inspired by ideas from evolutionary biology. However modern Evolutionary Computation (EC) only in outline reminds the strategies of biological evolution. The application of other algorithms and biological ideas may substantially improve the performance of this area of computer science. Namely, the selfish (or parasitic) mobile genetic elements - transposons are good candidates for this breakthrough. These genomic parasites live on a substratum of genomes of whole biological communities. Many biologists assume that processes in the world of transposons are the main source of evolution creativity. They thought to act as wise higher-level mutators for their hosts. In this communication we propose a strategy of construction of a new approach exploiting the most essential aspects of co-evolution of the hosts-chromosomes and their genetic parasites. We named this strategy as the Two-level Evolving Worlds. The key feature of the approach is usage of artificial transposons. We apply it to one of known benchmark problems - the John Muir ant's trail test. We found that our enhancement of GA technique by the artificial transposons obviously increase the efficacy of searching of the ant's navigation algorithm. We investigate in details the way of the transposons action as intelligent mutators of host-chromosomes.