지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술저널
- 저자정보
- 저널정보
- Korean Institute of Information Scientists and Engineers Journal of Electrical Engineering and Information Science Journal of Electrical Engineering and Information Science Vol.3 No.5
- 발행연도
- 1998.10
- 수록면
- 666 - 671 (6page)
이용수
초록· 키워드
오류제보하기
The Error Back-Propagation(EBP) algorithm is widely applied to train a multi-layer perceptron, which is a neural network model frequently employed in order to solve complex problems including pattern recognition and adaptive control. However, it suffers from two major problems: local minima and network structure design.
This paper presents a modified error back-propagation algorithm which have the capability to solve the local minima problem and the network structure design in a unified and efficient way. Our algorithm is basically the same as the conventional EBP algorithm except application of stochastic perturbation in order to escape a local minimum. In our algorithm when a local minimum is detected weights associated with hidden units are probabilistically reinitialized and the normal EBP learning is continued with the new set of weights. Addition of a new hidden unit also can be viewed as a special case of stochastic perturbation, i.e., reinitializing all-zero weights of a virtually existing unit. The results of our experiments with several benchmark test problems, the parity problem and the two-spirals problem, including "credit-screening" data, a practical problem of credit card approval, demonstrate that our algorithm is very efficient.
This paper presents a modified error back-propagation algorithm which have the capability to solve the local minima problem and the network structure design in a unified and efficient way. Our algorithm is basically the same as the conventional EBP algorithm except application of stochastic perturbation in order to escape a local minimum. In our algorithm when a local minimum is detected weights associated with hidden units are probabilistically reinitialized and the normal EBP learning is continued with the new set of weights. Addition of a new hidden unit also can be viewed as a special case of stochastic perturbation, i.e., reinitializing all-zero weights of a virtually existing unit. The results of our experiments with several benchmark test problems, the parity problem and the two-spirals problem, including "credit-screening" data, a practical problem of credit card approval, demonstrate that our algorithm is very efficient.
목차
Abstract
Ⅰ. Introduction
Ⅱ. The EBP/SP Learning Algorithm
Ⅲ. Simulation Results and Discussions
Ⅳ. Conclusions
Acknowledgement
References
저자소개
참고문헌 (0)
참고문헌 신청함께 읽어보면 좋을 논문
논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!
-
신경망 학습에서 국부적 최적해 탈출을 위한 확률적 교환
정보과학회논문지(B)
1998 .01
-
Weight Perturbation for Efficient Learning of Neural Networks
Journal of Electrical Engineering and information Science
1998 .10
-
역전파 알고리즘을 이용한 경계결정의 구성에 관한 연구
정보학연구
2005 .01
-
가속화 알고리즘을 이용한 EBP의 학습 속도의 개선에 관한 연구
대한전기학회 학술대회 논문집
1989 .11
-
Active Learning Algorithm for Neural Networks
ICONIP : International Conference On Neural Information Processing
1994 .01
-
가변학습율과 온라인모드를 이용한 개선된 EBP 알고리즘
한국정보과학회 학술발표논문집
2004 .04
-
ADAPTIVE CONTROL USING NEURAL NETWORK FOR MINIMUM-PHASE STOCHASTIC NONLINEAR SYSTEM
제어로봇시스템학회 국내학술대회 논문집
2000 .10
-
ADAPTIVE CONTROL USING NEURAL NETWORK FOR MINIMUM-PHASE STOCHASTIC NONLINEAR SYSTEM
제어로봇시스템학회 국내학술대회 논문집
2000 .10
-
A New Neural Network Algorithm to Solve the Optimal Problems
대한전자공학회 세미나
1994 .01
-
패리티 판별을 위한 유전자 알고리즘을 사용한 신경회로망의 학습법
전자공학회논문지-CI
2007 .03
-
Molecular Analysis of C/EBPβ and C/EBPδ Genes and Their Role After Exercise in Thoroughbred Horses
동물자원연구
2018 .01
-
Neural Network Structure Design Using Genetic Algorithm
제어로봇시스템학회 국내학술대회 논문집
1995 .10
-
Implementation of Self-adaptive System using the Algorithm of Neural Network Learning Gain
International Journal of Control, Automation, and Systems
2008 .06
-
개선된 역전파 신경회로망을 이용한 온라인 필기체 숫자의 분류에 관한 연구
정보학연구
2006 .01
-
Implementation of Self-adaptive System using the Algorithm of Neural Network Learning Gain
International Journal of Control Automation and Systems
2008 .06
-
A FILTERING CONDITION AND STOCHASTIC ADAPTIVE CONTROL USING NEURAL NETWORK FOR MINIMUM-PHASE STOCHASTIC NONLINEAR SYSTEM
대한전기학회 학술대회 논문집
2001 .11
-
신경회로망 학습이득 알고리즘을 이용한 자율적응 시스템 구현
대한전기학회 학술대회 논문집
2006 .07
-
Further Results for Linear Systems with Structured Perturbations
JTC-CSCC : Joint Technical Conference on Circuits Systems, Computers and Communications
1995 .01
이 논문의 저자 정보
최근 본 자료
댓글(0)
0
UCI(KEPA) : I410-ECN-0101-2009-569-017766502