Title

Evolutionary computing on consumer graphics hardware

Document Type

Journal article

Source Publication

IEEE Intelligent Systems

Publication Date

1-1-2007

Volume

22

Issue

2

First Page

69

Last Page

78

Abstract

Evolutionary algorithms (EA) are proven effective and robust in searching large varied spaces in a wide range of applications such as feature selection, electrical-circuit synthesis, and data mining. A growing demand from the multimedia and games industries has enabled graphics hardware companies to develop high-performance parallel graphics accelerators that resulted in the development of graphics processing unit (GPU). GPU handles rendering requests using a 3D graphics application programming interface (API). GPU lets processors communicate with any other processor directly, which enables to implement more flexible, fine grained EAs. A parallel EA can be implemented on consumer graphics cards found in many PCs. Evolutionary programming and genetic algorithms have been successfully applied to several numerical and optimization problems. EP requires mutation and is less computationally intensive than Genetic algorithm.

DOI

10.1109/MIS.2007.28

Print ISSN

15411672

E-ISSN

19411294

Publisher Statement

Copyright © 2007 Institute of Electrical and Electronics Engineers

Access to external full text or publisher's version may require subscription.

Full-text Version

Publisher’s Version

Recommended Citation

Fok, K.-L., Wong, T.-T., & Wong, M.-L. (2007). Evolutionary computing on consumer graphics hardware. IEEE Intelligent Systems, 22(2), 69-78. doi: 10.1109/MIS.2007.28