Abstract
In recent years we have been witness to the application of bio-inspired algorithms for the solution of a plethora of hard problems in computer science, engineering, biology, and so forth.1 Arguably the most popular of these bio-inspired methodologies is the evolutionary algorithm. Based on—and inspired by—the workings of evolution by natural selection, the basic meta-algorithm is seductively (and, I might add, deceptively) simple, and can be expressed in a mere eight lines of pseudocode:
1. produce an initial population of individuals, these latter being candidate solutions to the problem at hand
2. evaluate the fitness of each individual in accordance with the problem whose
solution is sought
3. while termination condition not met do
4. select fitter individuals for reproduction
5. recombine (crossover) individuals
6. mutate individuals
7. evaluate fitness of modified individuals
8. end while
1. produce an initial population of individuals, these latter being candidate solutions to the problem at hand
2. evaluate the fitness of each individual in accordance with the problem whose
solution is sought
3. while termination condition not met do
4. select fitter individuals for reproduction
5. recombine (crossover) individuals
6. mutate individuals
7. evaluate fitness of modified individuals
8. end while
| Original language | English |
|---|---|
| Pages (from-to) | 3 |
| Journal | AISB Quarterly: newsletter of the Society for the Study of Artificial Intelligence & Simulation of Behaviour |
| Issue number | 118 |
| State | Published - 2004 |