• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.8
• /
• pp.2119-2132
• /
• 1996

This paper proposes four approaches for approximately realizing nonlinear mappling of interval vectors by neural networks. In the proposed approaches, training data for the learning of neural networks are the paris of interval input vectors and interval target output vectors. The first approach is a direct application of the standard BP (Back-Propagation) algorithm with a pre-processed training data. The second approach is an application of the two BP algorithms. The third approach is an extension of the BP algorithm to the case of interval input-output data. The last approach is an extension of the third approach to neural network with interval weights and interval biases. These approaches are compared with one another by computer simulations.

• Korean Journal of Cognitive Science
• /
• v.4 no.2
• /
• pp.181-200
• /
• 1994

In the learning phase of multilyer feedforword neural network,there are problems such that local minimum,learning praralysis and slow learning speed when backpropagation algorithm used.To overcome these problems, the genetic algorithm has been used as learing method in the multilayer feedforword neural network instead of backpropagation algorithm.However,because the genetic algorith, does not have any mechanism for fine-tuned local search used in backpropagation method,it takes more time that the genetic algorithm converges to a global optimal solution.In this paper,we suggest a new GA-BP method which provides a fine-tunes local search to the genetic algorithm.GA-BP method uses gradient descent method as one of genetic algorithm's operators such as mutation or crossover.To show the effciency of the developed method,we applied it to the 3-parity bit problem with analysis.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.477-480
• /
• 2005

본 논문에서는 신경망과 유전자 알고리즘의 장점을 결합하고, 개선된 유전자 알고리즘 기반의 역전파 신경망 알고리즘을 이용한 신경망 학습 효과를 살펴 본다. 유전자 알고리즘을 이용한 신경망 학습은 비선형 함수를 이용하여 발생시킨 모의 자료를 통하여 수행하고 학습 수렴의 정도와 학습 속도 등을 비교할 수 있는 모의실험 결과를 일반 신경망 학습 결과와 함께 제시한다. 모의실험의 결과로서 유전자 알고리즘을 적용한 신경망 제어기가 일반 신경망 학습 결과보다 수렴 정확도 및 학습 속도에서 더 좋은 결과를 나타내 주고 있다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.4
• /
• pp.883-889
• /
• 2005

Error backpropagation algorithm of multilayer perceptron may result in local-minima because of the insufficient nodes in the hidden layer, inadequate momentum set-up, and initial weights. In this paper, we proposed the ART-1 based fuzzy supervised learning algorithm which is composed of ART-1 and fuzzy single layer supervised learning algorithm. The Proposed fuzzy supervised learning algorithm using self-generation method applied not only ART-1 to creation of nodes from the input layer to the hidden layer, but also the winer-take-all method, modifying stored patterns according to specific patterns. to adjustment of weights. We have applied the proposed learning method to the problem of recognizing a resident registration number in resident cards. Our experimental result showed that the possibility of local-minima was decreased and the teaming speed and the paralysis were improved more than the conventional error backpropagation algorithm.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.5
• /
• pp.1393-1402
• /
• 1999

This paper proposes an efficient principal feature extraction of the image data using neural networks of a new learning algorithm. The proposed learning algorithm is a backpropagation(BP) algorithm based on the steepest descent and dynamic tunneling. The BP algorithm based on the steepest descent is applied for high-speed optimization, and the BP algorithm based on the dynamic tunneling is also applied for global optimization. Converging to the local minimum by the BP algorithm of steepest descent, the new initial weights for escaping the local minimum is estimated by the BP algorithm of dynamic tunneling. The proposed algorithm has been applied to the 3 image data of 12${\times}$12pixels and the Lenna image of 128${\times}$128 pixels respectively. The simulation results shows that the proposed algorithm has better performances of the convergence and the feature extraction, in comparison with those using the Sanger method and the Foldiak method for single-layer neural networks and the BP algorithm for multilayer neural network.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 1998.04a
• /
• pp.135-140
• /
• 1998

본 논문에서는 신경회로망에서 가장 널리 쓰이고 있는 오차 역 전파 알고리즘(Error Back-propagation) 을 사용하여 스캐너를 모델링함으로써 스캐너의 원색 재현을 위한 방법을 제시하였다. 이것은 스캐너의 하드웨어적 특성을 고려, 입력된 영상의 원색과 출력물의 색과 일치시키는 방법이다. 우선, 오차 역전파 알고리즘에 대하여 학습 규칙을 살펴보고 학습을 위한 데이터를 추출하기 위해 고르게 분포된 색 샘플들을 계측기로 측정하여 칼라 공간에서의 X, Y, Z 값을 얻어낸다. 그 중에서 표본 샘플을 추출한다. 그리고 이를 스캐너로 스캐닝하여 얻은 R, G, B값을 오차 역전파 알고리즘의 입력값으로, 목표값은 X, Y, Z값을 사용하여 학습시킨다. 학습하는 동안 샘플 색상의 수와 중간층의 수, 노드의 수를 변화시킴으로써 최적의 결과를 얻도록 실험하였다. 결론에서는 서로간의 결과를 분석한다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.11b
• /
• pp.129-132
• /
• 2002

본 논문은 웨이브렛 변환 다중해상도 분해(multi-resolution Analysis : MRA)와 적응성 뉴로-퍼지 인터페이스 시스템(Adaptive Neuro-Fuzzy Inference System : ANFIS)을 기반으로 한 웨이브렛 신경망을 가지고 임의의 비선형 함수 학습 근사화를 개선하는 것이다. ANFIS 구조는 벨형 퍼지 함수로 구성이 되었고, 웨이브렛 신경망은 전파 알고리즘과 역전파 신경망 알고리즘으로 구성되었다. 여기 웨이브렛 구성은 단일 크기이고, ANFIS 기반 웨이브렛 신경망의 학습을 위해 역전파 알고리즘을 사용하였다. 1차원과 2차원 함수에서 웨이브렛 전달 파라미터 학습과 ANFIS의 벨형 소속 함수를 이용한 ANFIS 모델 기반 웨이브렛 신경망의 웨이브렛 기저 수 감소와 수렴 속도 성능이 기존의 알고리즘 보다 개선되었음을 확인하였다.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.163-171
• /
• 1997

Function approximation from a set of input-output pairs has numerous applications in scientific and engineering areas. Multilayer feedforward neural networks have been proposed as a good approximator of nonlinear function. The back propagation(BP) algorithm allows multilayer feedforward neural networks to learn input-output mappings from training samples. It iteratively adjusts the network parameters(weights) to minimize the sum of squared approximation errors using a gradient descent technique. However, the mapping acquired through the BP algorithm may be corrupt when errorneous training data we employed. When errorneous traning data are employed, the learned mapping can oscillate badly between data points. In this paper we propose a robust BP learning algorithm that is resistant to the errorneous data and is capable of rejecting gross errors during the approximation process, that is stable under small noise perturbation and robust against gross errors.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.1
• /
• pp.21-31
• /
• 2000

In this paper, an analysis of the stability for a class of discrete-time neural network control systems is presentd. Based on Lyapunov's direct method, a sufficient stability condition for the neural network control systems is systematically derived and the modified back propagation algorithm which reflects the derived stability condition is suggested. The modified BP originates from the derived sufficient condition and guarantees the exponential stability of the resulting trained closed system. Finally, computer simulation is included to show an example where the derived stability condition and the BP modified bythe condition is used to train the control plant.

• Journal of Intelligence and Information Systems
• /
• v.16 no.3
• /
• pp.1-14
• /
• 2010

The back-propagation neural network (BPN) has long been successfully applied in bankruptcy prediction problems. Despite its wide application, some major issues must be considered before its use, such as the network topology, learning parameters and normalization methods for the input and output vectors. Previous studies on bankruptcy prediction with BPN have shown that many researchers are interested in how to optimize the network topology and learning parameters to improve the prediction performance. In many cases, however, the benefits of data normalization are often overlooked. In this study, a genetic algorithm (GA)-based normalization transform, which is defined as a linearly weighted combination of several different normalization transforms, will be proposed. GA is used to extract the optimal weight for the generalization. From the results of an experiment, the proposed method was evaluated and compared with other methods to demonstrate the advantage of the proposed method.