Evolutionary algorithms are stochastic optimization methods that iteratively improve a population of solutions in a search space through principles of natural selection and variation.
Formally, EAs include operators such as selection, crossover (recombination), and mutation to generate new individuals, which are evaluated based on a fitness function.
The algorithm repeats these steps over generations until a termination criterion is met.
Evolutionary algorithms are particularly effective for multimodal, non-differentiable, or discrete optimization problems.
Definition:
“Evolutionary algorithms are population-based metaheuristics that optimize a set of solutions through iterative application of selection, recombination, and mutation, inspired by biological evolution.”
Source:
Back, T., Fogel, D. B., & Michalewicz, Z. (Eds.). (2018). Handbook of Evolutionary Computation. IOP Publishing.