• DocumentCode
    2322649
  • Title

    Stress Calculate: Stress Reaction System Model

  • Author

    Gu, Wen-xiang ; Liu, Yan ; Jin, Xi-Zi

  • Author_Institution
    Computer department of northeast normal university, changchun, china; E-MAIL: gwx@nenu.edu.cn
  • Volume
    1
  • fYear
    2005
  • fDate
    18-21 Aug. 2005
  • Firstpage
    221
  • Lastpage
    224
  • Abstract
    Enemy (competition opponent) execute a plan, which has harmful goals for us, this plan is called adversarial plan (opponent plan). To do the answer for the adversarial plan (opponent plan) is called reply plan. The start mechanism of reply plan is the main problem in this paper. In the paper, we introduce a new model: Stress Reaction System Model. We also give the definitions of stress reaction and stress calculation algorithms in this paper. This model has the property of rapidly reaction. When the intelligent system (based on the model of stress reaction system) faces an adversarial attack, it can learn and remember. Because of this, it can defend or counter-attack when meets the attack which it had met before without recognize again. Inhibit chain is used to control the reaction, prevent excessive reaction (sometimes, excessive reaction make reverse influence). Stress reaction system model research is very important for information security, robot competition, play-role selection and so on.
  • Keywords
    Stress calculation algorithm; inhibit chain; learning and memory model based on immunity; stress reaction; synapse transfer; Artificial intelligence; Biological system modeling; Face recognition; Immune system; Information security; Intelligent systems; Organisms; Robots; Sequences; Stress; Stress calculation algorithm; inhibit chain; learning and memory model based on immunity; stress reaction; synapse transfer;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
  • Conference_Location
    Guangzhou, China
  • Print_ISBN
    0-7803-9091-1
  • Type

    conf

  • DOI
    10.1109/ICMLC.2005.1526948
  • Filename
    1526948