GPGPGPU: Evaluation of Parallelisation of Genetic Programming using GPGPU

Created by W.Langdon from gp-bibliography.bib Revision:1.8178

@InProceedings{Kim:2017:SSBSE,

• author = "Jinhan Kim and Junhwi Kim and Shin Yoo",
• title = "{GPGPGPU}: Evaluation of Parallelisation of Genetic Programming using GPGPU",
• booktitle = "Proceedings of the 9th International Symposium on Search Based Software Engineering, SSBSE 2017",
• year = "2017",
• editor = "Tim Menzies and Justyna Petke",
• volume = "10452",
• series = "LNCS",
• pages = "137--142",
• address = "Paderborn, Germany",
• month = sep # " 9-11",
• publisher = "Springer",
• keywords = "genetic algorithms, genetic programming, GPU",
• isbn13 = "978-3-319-66299-2",
• DOI = "doi:10.1007/978-3-319-66299-2_11",
• size = "6 pages",
• abstract = "We evaluate different approaches towards parallelisation of Genetic Programming (GP) using General Purpose Computing on Graphics Processor Units (GPGPU). Unlike Genetic Algorithms, which uses a single or a fixed number of fitness functions, GP has to evaluate a diverse population of programs. Since GPGPU is based on the Single Instruction Multiple Data (SIMD) architecture, parallelisation of GP using GPGPU allows multiple approaches. We study three different parallelisation approaches: kernel per individual, kernel per generation, and kernel interpreter. The results of the empirical study using a widely studied symbolic regression benchmark show that no single approach is the best: the decision about parallelisation approach has to consider the trade-off between the compilation and the execution overhead of GPU kernels.",
• notes = "Short Papers http://ssbse17.github.io/ Co-located with FSE/ESEC 2017",

}

Genetic Programming entries for Jinhan Kim Junhwi Kim Shin Yoo

Citations