Algorithm Implementation

This section describes more specifically how Evolver’s Genetic Algorithm optimization is implemented.

Please note: Knowledge of this material is not necessary to use Evolver.

The majority of Evolver’s genetic algorithm technology such as the Recipe and Order solving methods are based on academic work in the genetic algorithm field over the last decade or two. However, most of the descendant solving methods included with Evolver, and the multiple groups of adjustable cells, backtracking, strategy, and probability features are unique to Evolver.

Evolver uses a steady-state approach. This means that only one organism is replaced at a time, rather than an entire “generation” being replaced. This steady-state technique has been shown to work as well or better than the generational replacement method. To determine the equivalent number of “generations” Evolver has run, divide the number of individual trials it has explored by the size of the population.