선형계획 문제의 해를 구하는 신경회로

Neural Networks for Solving Linear Programming Problems and Linear Systems

  • 장석호 (강원대학교 전자공학과) ;
  • 강성귀 (강원대학교 전자공학과) ;
  • 남부희 (강원대학교 제어계측공학과) ;
  • 이정문 (강원대학교 제어계측공학과)
  • Chang, S.H. (Dept of Electronics Eng., Kangwon National University) ;
  • Kang, S.G. (Dept of Electronics Eng., Kangwon National University) ;
  • Nam, B.H. (Dept of Control and Instrumentation Eng. Kangwon National University) ;
  • Lee, J.M. (Dept of Control and Instrumentation Eng. Kangwon National University)
  • 발행 : 1993.07.18

초록

The Hopfield model is defined as an adaptive dynamic system. In this paper we propose a modified neural network which is capable of solving linear programming problems and a set of linear equations. The model is directly implemented from the given system, and solves the problem without calculating the inverse of the matrices. We get the better stability results by the addition of scaling property and by using the nonlinearities in the linear programming neural networks.

키워드