- author = "Petr Pospichal and Eoin Murphy and Michael O'Neill and Josef Schwarz and Jiri Jaros",
- title = "Acceleration of grammatical evolution using graphics processing units: computational intelligence on consumer games and graphics hardware",
- booktitle = "GECCO 2011 Computational intelligence on consumer games and graphics hardware (CIGPU)",
- year = "2011",
- editor = "Simon Harding and W. B. Langdon and Man Leung Wong and Garnett Wilson and Tony Lewis",
- isbn13 = "978-1-4503-0690-4",
- keywords = "genetic algorithms, genetic programming, grammatical evolution, GPU, CUDA, GPGPU, symbolic regression",
- pages = "431--438",
- month = "12-16 " # jul,
- organisation = "SIGEVO",
- address = "Dublin, Ireland",
-
DOI = "
doi:10.1145/2001858.2002030",
- publisher = "ACM",
- publisher_address = "New York, NY, USA",
-
abstract = "Several papers show that symbolic regression is
suitable for data analysis and prediction in financial
markets. Grammatical Evolution (GE), a grammar-based
form of Genetic Programming (GP), has been successfully
applied in solving various tasks including symbolic
regression. However, often the computational effort to
calculate the fitness of a solution in GP can limit the
area of possible application and/or the extent of
experimentation undertaken. This paper deals with using
mainstream graphics processing units (GPU) for
acceleration of GE solving symbolic regression. GPU
optimisation details are discussed and the NVCC
compiler is analysed. We design an effective mapping of
the algorithm to the CUDA framework, and in so doing
must tackle constraints of the GPU approach, such as
the PCI-express bottleneck and main memory
transactions.
This is the first occasion GE has been adapted for running on a GPU. We measure our implementation running on one core of CPU Core i7 and GPU GTX 480 together with a GE library written in JAVA, GEVA.
Results indicate that our algorithm offers the same convergence, and it is suitable for a larger number of regression points where GPU is able to reach speedups of up to 39 times faster when compared to GEVA on a serial CPU code written in C. In conclusion, properly used, GPU can offer an interesting performance boost for GE tackling symbolic regression.",
-
notes = "Two kernels: 1) selection 2) genotype-phenotype
mapping, fitness, crossover, mutation. One grammar for
problem x +x^2 + x^3 + x^4. Tries Population 2 ... 64
and 128 ... 2560 training cases. Comparison with PC in
C and Java (GEVA). No absolute speed measure given (cf.
\cite{langdon:2008:eurogp}).
Also known as \cite{2002030} Distributed on CD-ROM at GECCO-2011.
ACM Order Number 910112.",
