Evolution Strategies (ESs) are in many ways very similar to Genetic Algorithms (GAs). As their name implies, ESs too simulate natural evolution. The differences between GAs and ESs arise primarily because the original applications for which the algorithms were developed are different. While GAs were designed to solve discrete or integer optimization problems, ESs were applied first to continuous parameter optimization problems associated with laboratory experiments.

ESs were introduced in the 1960s by Rechenberg [65] working in Berlin and further developed by Schwefel [66]. The first numerical simulations were performed by Hartmann [64], and the first attempts at using ESs to solve discrete optimization were made by Schwefel [67].

Like GAs, ESs differ from traditional optimization algorithms in some important respects:

- They search from one population of solutions to another, rather than
from individual to individual.
- They use only objective function information, not derivatives.
- They use probabilistic, not deterministic, transition rules.