References
- 강원대학교 논문집 v.35 Mathematica를 이용한 고정점 반복법 박기성
- Mathematica와 미분적분학 신창언
- 패턴인식의 원리 이성환
- Simulating Neural Networks with Mathematica Freeman, James A.
- Mathematica: A System for Doing Mathematics by Computer Wolfram, S.
Journal of the Korean Data and Information Science Society
- Volume 16 Issue 4
- /
- Pages.815-821
- /
- 2005
- /
- 1598-9402(pISSN)
Improved Learning Algorithm with Variable Activating Functions
Abstract
Among the various artificial neural networks the backpropagation network (BPN) has become a standard one. One of the components in a neural network is an activating function or a transfer function of which a representative function is a sigmoid. We have discovered that by updating the slope parameter of a sigmoid function simultaneous with the weights could improve performance of a BPN.