Convergence (evolutionary computing)

Convergence is a phenomenon in evolutionary computation. It causes evolution to halt because precisely every individual in the population is identical. Full convergence might be seen in genetic algorithms (a type of evolutionary computation) using only crossover (a way of combining individuals to make new offspring). Premature convergence is when a population has converged to a single solution, but that solution is not as high of quality as expected, i.e. the population has gotten 'stuck'. However, convergence is not necessarily a negative thing, because populations often stabilise after a time, in the sense that the best programs all have a common ancestor and their behaviour is very similar (or identical) both to each other and to that of high fitness programs from the previous generations. Often the term convergence is loosely used. Convergence can be avoided with a variety of diversity-generating techniques.

References

Foundations of Genetic Programming

Evolutionary computation

Main Topics	Convergence (evolutionary computing) Evolutionary algorithm Evolutionary data mining Evolutionary multimodal optimization Human-based evolutionary computation Interactive evolutionary computation

Algorithms	Cellular evolutionary algorithm Covariance Matrix Adaptation Evolution Strategy (CMA-ES) Differential evolution Evolutionary programming Genetic algorithm Genetic programming Gene expression programming Evolution strategy Natural evolution strategy Neuroevolution Learning classifier system

Related topics	Artificial development Artificial intelligence Developmental biology Digital organism Evolutionary robotics Fitness function Fitness landscape Fitness approximation Genetic operators Interactive evolutionary computation List of digital organism simulators No free lunch in search and optimization Machine learning Program synthesis

Journals	Evolutionary Computation (journal)

This article is issued from Wikipedia - version of the 3/21/2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.