Efficient Use of Variation in Evolutionary Optimization

Evolutionary algorithms face a fundamental trade-off between exploration and exploitation. Rapid performance improvementtends to be accompanied by a rapid loss of diversity from the population of potential solutions, causing premature convergenceon local rather than global optima. However, the rate at which diversity is lost from a population is not simply a function of thestrength of selection but also its e fficiency, or rate of performance improvement relative to loss of variation. Selection e fficiencycan be quantified as the linear correlation between objective performance and reproduction. Commonly used selection algorithmscontain several sources of ine fficiency, some of which are easily avoided and others of which are not. Selection algorithms basedon continuously varying generation time instead of discretely varying number of off spring can approach the theoretical limit onthe e fficient use of population diversity.