Natural Computation

From NECSIWiki

Natural Computation
Simulated Annealing Genetic Algorithms Artificial Neural Networks Particle Swarm Optimization Ant Colony Optimization where is it? Reinforcement Learning Game Theory Probability Collectives Linking Natural Computation with Complex Systems edit box

Natural computation refers to the study of computational systems that use ideas and are inspired by biological, ecological and physical systems. More broadly this includes Quantum Computation and Robotics, although here we concentrate on applications in Optimisation. Natural phenomena have been an important inspiration for new methods in optimisation since John Holland developed the genetic algorithm in 1975. Techniques such as simulated annealing, artificial neural networks particle swarm optimisation and ant colony optimisation are dispersed across many disciplines including mathematics, computer science, engineering, operations research, cognitive science and artificial life. One field that is broad enough to encompass all of these models of optimisation is complex systems. Here we survey the available methods as seen from the field of complex systems. Each method is introduced with the original motivation, and then abstracted to a bare bones mathematical model. This uncovers relationships between techniques that on the surface appear unrelated. Finally, two important questions are addressed: why do natural complex systems provide good models for optimisation, and how do naturally inspired optimisation techniques enable a deeper understanding of complex systems.