Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지

Evolutionary Learning Algorithm fo r Projection Neural NEtworks

투영신경회로망의 훈련을 위한 진화학습기법

  • Published : 1997.10.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper proposes an evolutionary learning algorithm to discipline the projection neural nctworks (PNNs) with special type of hidden nodes which can activate radial basis functions as well as sigmoid functions. The proposed algorithm not only trains the parameters and the connection weights hut also c~ptimizes the network structure. Through the structure optimization, the number of hidden node:; necessary to represent a given target function is determined and the role of each hidden node is decided whether it activates a radial basis function or a sigmoid function. To apply the algorithm, PNN is realized by a self-organizing genotype representation with a linked list data structure. Simulations show that the algorithm can build the PNN with less hidden nodes than thc existing learning algorithm using error hack propagation(EE3P) and network growing strategy.

본 논문에서는 시그모이드 함수와 방사형 기저 함수 모두를 생성시킬 수 있는 특별한 은닉층 노드를 갖는 투영신경회로망에 대하여 알아롭고 그것을 훈련시키기 위한 진화 학습 기법을 제시한다. 제시된 기법은 신경회로망의 매개변수와 연결 가충치뿐만 아니라, 어떤 목적함수를 나타내기 위한 최적의 은닉층 노드개수 또한 구조 최적화를 위한 진화연산자를 통해 찾아낸다. 각각의 은닉층 노드의 역할은 진화를 거듭하면서 방사형 기저 함수를 나타낼지 시그모이드 함수를 나타낼지 결정된다. 알고리즘을 구현하기 위해서 투영신경회로망은 연결 고리 리스트 자료구조로 나타내었다. 모의 실험에서 기존으 오차역전파에 의한 학습과 구조 성장 방식보다 적은 노드로 투영신경회로망을 훈련시킬 수 있음을 볼수 있다.

Keywords

References

  1. Neural Network Design and the Complexity fo Learning J.S.Judd
  2. Representation and Evolution of Neural Networks Martin Mandischer
  3. Procs. of IJCNN v.1 A hybrid model composed of a multilayer perceptron and a radial basis function network M.Hirahara;N.Oka
  4. WCNN v.4 Time series prediction based on global and local estimation models R.M.Kil;J.Y.Choi
  5. Procs. of ICONIP v.3 Structural hybrid of multilayer neural networks and Gaussian potential function netwoks S.Park;L.J.Park;C.H.Park
  6. Journal of Korean Institute of Communication Sicences v.18 no.12 Efficient learning algorithm using structural hybrid of multilayer neural networks and Gaussian ptential function networks S.Park;L.J.Park;C.H.Park
  7. KIEE v.44 Projection Neural Networks for Function Approximation Byeong-Kap Choi;Chong-Ho Choi;Jin-Young Choi
  8. Genetic Algorithms in Search, Optimization and Machine Learning D.E.Goldberg
  9. Proceedings of the Third Annual Conference on Evolutionary Programming Evolutionary Operators for Continuous Convex Parameter Spaces Z.Michalewicz;T.D.Logan;S.Swaminathan
  10. Heuristical Constraints on Search Complexity for Multi-modal Non-optimal Models Mukesh J. Patel
  11. Procs. of IJCNN v.2 The Projection neural networks G.D.Wilensky;N.Manukian
  12. IEEE Transactions on Neural Networks v.5 no.1 Convergence analysis of canonical genetic algorithms Gunter Rudolph