Genetic Algorithms
This is a group of algorithms that aim at solving Optimization problems with ideas borrowed from biology (evolutions)
General idea:
- start with a population of potential solutions
- score each solution
- select the best (only the best “survive”)
- perform some random mutations (permutations, etc)
- combine the solutions in hope to get a better one (cross-over)
- repeat for several generations
Applications
This model is successfully applied in many domains: