Special Issue ...

Evolutionary Computation On General Purpose Graphics Processing Units


Soft Computing - A Fusion of Foundations, Methodologies and Applications

Springer - ISSN: 1432-7643 (print version) ISSN: 1433-7479 (electronic version)

Special issue on Evolutionary Computation on General Purpose Graphics Processing Units


Only papers with minor review required will be accepted

Guest Editors:

Associate Editor:

We are very pleased to invite you to submit your paper for consideration for a special issue of the Soft Computing journal: Evolutionary Computation on General Purpose Graphics Processing Units.

Evolutionary algorithms (EAs) are metaheuristics, inspired by natural phenomena in a broader sense, that have proven to be effective for difficult combinatorial optimization problems appearing in various industrial, economical, and scientific domains. Prominent examples for such problems are network design, packing, satisfiability, scheduling, timetabling, transportation, traveling salesperson, vehicle routing, circuits design, to name but a few. Given the large amount of computational resources frequently required for solving hard problems, ideas taken from parallel computing has been successfully applied to different evolutionary algorithms during the last few years, such as Simulated Annealing, Genetic Algorithms, Ant Colony Optimization, Immune Systems, Genetic Programming, etc.

In this regard, the use of Graphics Processing Units (GPUs) in scientific computing is becoming increasingly common. GPUs are low cost parallel processors that can readily be exploited for many types of general purpose computation. Recently, the computational intelligence community has started to develop for the GPU platform. With the advances of modern consumer-level GPUs, parallel EAs can be designed, that fits on Single Instruction, Multiple Data (SIMD)-based GPU. With the low-cost GPU equipped on ordinary PC, more people will be able to use parallel evolutionary algorithms to solve huge problems encountered in real-world applications. There are many interesting issues for future work, for example, current generation of GPU present some issues to generate random numbers because these random numbers will lead to poor performance of EAs. This special issue particularly focuses on the applications of evolutionary computation that can maximally exploit the parallelism provided by graphics processing units.

Authors are invited to submit their original and unpublished work (theoretical and empirical contributions, as well as surveys) in the areas of Evolutionary Computation on GPGPUs.

Manuscripts must follow author guidelines for the journal specified in (http://www.springer.com/engineering/journal/500) and must be submitted by email to jlrisco@dacya.ucm.es indicating - Special Issue ECGPU - by the closing date (December 5, 2010). All submitted papers will be double blind reviewed. Selection criteria will be based on relevance, originality, significance, impact, technical soundness and quality of the paper. Contributions are expected to provide original results, insights and experimental innovations.

Please contact us if you need further information

With best wishes,

Yours sincerely,

José L. Risco-Martín jlrisco@dacya.ucm.es

Juan Lanchares julandan@dacya.ucm.es

Guest Editors