• DocumentCode
    1797385
  • Title

    Cognitive neural network for cybersecurity

  • Author

    Perlovsky, Leonid ; Shevchenko, Olexander

  • Author_Institution
    Athinoula Martinos Imaging Center, Harvard Univ., Charlestown, MA, USA
  • fYear
    2014
  • fDate
    6-11 July 2014
  • Firstpage
    4056
  • Lastpage
    4061
  • Abstract
    This chapter discusses the future of cybersecurity as warfare between machine learning techniques of attackers and defenders. As attackers will learn to evolve new camouflaging methods for evading better and better defenses, defense techniques will in turn learn new attacker´s tricks to defend against. The better technology will win. Here we discuss theory of machine learning based on dynamic logic that are mathematically provable to learn with the fastest possible speed. We also discuss cognitive functions of dynamic logic and related experimental proofs. This new mathematical theory, in addition to being provably fastest machine learning technique, is also an adequate model for several fundamental mechanisms of the mind.
  • Keywords
    cognitive systems; learning (artificial intelligence); neural nets; security of data; camouflaging methods; cognitive functions; cognitive neural network; cybersecurity; dynamic logic; experimental proofs; machine learning techniques; Abstracts; Cognition; Computers; Machine learning algorithms; Malware; Mathematical model;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks (IJCNN), 2014 International Joint Conference on
  • Conference_Location
    Beijing
  • Print_ISBN
    978-1-4799-6627-1
  • Type

    conf

  • DOI
    10.1109/IJCNN.2014.6889430
  • Filename
    6889430