Publication detail

Evolutionary Approximation of Software for Embedded Systems: Median Function

MRÁZEK, V. VAŠÍČEK, Z. SEKANINA, L.

Original Title

Evolutionary Approximation of Software for Embedded Systems: Median Function

Type

conference paper

Language

English

Original Abstract

This paper deals with genetic programming-based improvement of non-functional properties of programs intended for low-cost microcontrollers. As the objective is to significantly reduce power consumption and execution time, the approximate computing scenario is considered in which occasional errors in results are acceptable. The method is based on Cartesian genetic programming and evaluated in the task of approximation of 9-input and 25-input median function. Resulting approximations show a significant improvement in the execution time and power consumption with respect to the accurate median function while the observed errors are moderate.

Keywords

Genetic Improvement; Genetic Programming; Cartesian Genetic Programming; Approximate Computing; Embedded Systems

Authors

MRÁZEK, V.; VAŠÍČEK, Z.; SEKANINA, L.

RIV year

2015

Released

3. 7. 2015

Publisher

Association for Computing Machinery

Location

New York

ISBN

978-1-4503-3488-4

Book

GECCO Companion '15 Proceedings of the Companion Publication of the 2015 on Genetic and Evolutionary Computation Conference

Edition

ACM

Pages from

795

Pages to

801

Pages count

7

URL

https://www.fit.vut.cz/research/publication/10852/

BibTex

@inproceedings{BUT119838,
  author="Vojtěch {Mrázek} and Zdeněk {Vašíček} and Lukáš {Sekanina}",
  title="Evolutionary Approximation of Software for Embedded Systems: Median Function",
  booktitle="GECCO Companion '15 Proceedings of the Companion Publication of the 2015 on Genetic and Evolutionary Computation Conference",
  year="2015",
  series="ACM",
  pages="795--801",
  publisher="Association for Computing Machinery",
  address="New York",
  doi="10.1145/2739482.2768416",
  isbn="978-1-4503-3488-4",
  url="https://www.fit.vut.cz/research/publication/10852/"
}