Bio-Inspired Models of Network, Information, and Computing Systems. 5th International ICST Conference, BIONETICS 2010, Boston, USA, December 1-3, 2010, Revised Selected Papers

Research Article

Program Equivalence Using Neural Networks

Download
434 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-32615-8_63,
        author={Tiago Nascimento and Charles Prado and Davidson Boccardo and Luiz Carmo and Raphael Machado},
        title={Program Equivalence Using Neural Networks},
        proceedings={Bio-Inspired Models of Network, Information, and Computing Systems. 5th International ICST Conference, BIONETICS 2010, Boston, USA, December 1-3, 2010, Revised Selected Papers},
        proceedings_a={BIONETICS},
        year={2012},
        month={10},
        keywords={},
        doi={10.1007/978-3-642-32615-8_63}
    }
    
  • Tiago Nascimento
    Charles Prado
    Davidson Boccardo
    Luiz Carmo
    Raphael Machado
    Year: 2012
    Program Equivalence Using Neural Networks
    BIONETICS
    Springer
    DOI: 10.1007/978-3-642-32615-8_63
Tiago Nascimento,*, Charles Prado1,*, Davidson Boccardo1,*, Luiz Carmo1,*, Raphael Machado1,*
  • 1: INMETRO - National Institute of Metrology
*Contact email: tmnascimento@inmetro.gov.br, cbprado@inmetro.gov.br, drboccardo@inmetro.gov.br, lfrust@inmetro.gov.br, rcmachado@inmetro.gov.br

Abstract

Program equivalence refers to the mapping between equivalent codes written in different languages – including high-level and low-level languages. In the present work, we propose a novel approach for correlating program codes of different languages using artificial neural networks and program characteristics derived from control flow graphs and call graphs. Our approach correlates the program codes of different languages by feeding the neural network with logical flow characteristics. Our evaluation using real code examples shows a typical correspondence rate between 62% and 100% with the very low rate of 4% false positives.