In computer science, an evolution strategy (ES) is an optimization technique based on ideas of evolution. It belongs to the general class of evolutionary computation or artificial evolution methodologies.

(ES) A kind of evolutionary algorithm where individuals (potential solutions) are encoded by a set of real-valued "object variables" (the individual's "genome"). For each object variable an individual also has a "strategy variable" which determines the degree of mutation to be applied to the corresponding object variable. The strategy variables also mutate, allowing the rate of mutation of the object variables to vary. An ES is characterised by the population size, the number of offspring produced in each generation and whether the new population is selected from parents and offspring or only from the offspring. ES were invented in 1963 by Ingo Rechenberg, Hans-Paul Schwefel at the Technical University of Berlin (TUB) while searching for the optimal shapes of bodies in a flow. (1995-02-03)

Natural evolution strategies (NES) are a family of numerical optimization algorithms for black box problems. Similar in spirit to evolution strategies, they iteratively update the (continuous) parameters of a search distribution by following the natural gradient towards higher expected fitness.