β Scribed by Albert Y. Zomaya; Stephan Olariu
No coin nor oath required. For personal study only.
Computational models based on natural phenomena have gained popularity in recent years. This new direction has been prompted by the need to find solutions for a wide range of formidable problems that the prevailing mode of thinking could not provide satisfactory solutions for.
A number of studies reported in the last few years have shown that nature-inspired techniques have a great potential for solving a wide range of problems and also for influencing the design of future computers [1]. Some of these techniques, such as artificial life techniques and neural networks, are now commonplace and have been accepted by the wider scientific community. These techniques are widely used to solve a variety of optimization problems. They tend to excel in cases where the knowledge space is ambiguous or incomplete.
Other techniques, such as DNA-based and quantum computations, are more esoteric in nature. These techniques are perceived to hold the promise of building more powerful computers that are massively parallel in nature and hence able to provide considerable computing power that is not available today. This will enable, of course, the solution of more difficult problems that are computationally intractable by today's standards.
Techniques based on evolutionary paradigms can provide efficient solutions to a wide variety of problems in parallel and distributed computing. A vast literature exists on the use of evolutionary computing approaches to solve a wide range of problems and, more recently, a number of studies have reported on the success of such techniques in solving difficult problems in key areas of parallel and distributed computing [2]. Rather remarkably, most evolutionary techniques are inherently
π SIMILAR VOLUMES