Earticle

초록

영어

This essay proposes a method of BP neural network constrained optimization based on previous studies. The optimization method based on the BP neural network, takes the minimum output of a neural network as an example, gives the general mathematical models, derives and gives the partial derivatives of BP network's output to input, and uses the Sigmoid Function as the transmission function in the article. On the previous studies basis, the basic ideas, algorithms and related models are given, based on the constrained optimization issues of BP neural network. We can adjust the input values of BP neural network to obtain the minimum or maximum output value by using this method. This optimization method links the optimization and fitting of BP networks together and expands the application of the BP neural network. At last, in this essay, the optimization method is applied in an example.

목차

Abstract
 1. Preface
 2. The BP Neural Network’s Structure and Algorithm
  2.1. The BP Neural Network’s Structure
  2.2. The BP Neural Network’s Algorithm
 3. The Method of Constrained Optimization
  3.1. Mathematical Types
  3.2. The Basic Ideas
  3.3. Calculation of Partial Derivative
  3.4. The Method of Constrained Optimization
 4. The Example Calculation
 5. Conclusion
 References

키워드

저자

• Shuang-jing Li [ Northeast Agricultural University, Harbin, China ]
• Fu-lin Wang [ Northeast Agricultural University, Harbin, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Hybrid Information Technology

• 간기

  격월간

• pISSN

  1738-9968

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605