This paper rpesents a systematic approach to co-evolution that allows concise and unified expression of all types of symbiotic relationships studied in ecology. The resulting Linear Model of Symbiosis can be easily added to any regular Genetic Algorithm. Our model helps prevent premature convergence to a local optimum by maintaining the genetic diversity in a population. Our experiments show that co-evolutionary Genetic Al-gorithms outperform regular Genetic Algorithms on some difficult problems including one (Holland's Royal Road function) which was specifically designed to highlight the strengths of a regular Genetic Algorithm.
展开▼