• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.6
• /
• pp.67-75
• /
• 1998

Although the nCE(n-th order cross-entropy) error function resolves the incorrect saturation problem of conventional EBP(error back-propagation) algorithm, the performance of MLP's (multilayer perceptrons) trained using the nCE function depends heavily on the order of the nCE function. In this paper, we propose an adaptive learning rate to make the performance of MLP's insensitive to the order of the nCE error. Additionally, we propose a limited error signal of output node to prevent unstable learning due to the adaptive learning rate. The effectiveness of the proposed method is demonstrated in simulations of handwritten digit recognition and thyroid diagnosis tasks.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.1159-1163
• /
• 2010

최근의 극심한 기상이변으로 인하여 발생되는 유출량의 예측에 관한 사항은 치수 이수는 물론 방재의 측면에서도 역시 매우 중요한 관심사로 부각되고 있다. 강우-유출 관계는 유역의 수많은 시 공간적 변수들에 의해 영향을 받기 때문에 매우 복잡하여 예측하기 힘든 요소이다. 과거에는 추계학적 예측모형이나 확정론적 예측모형 혹은 경험적 모형 등을 사용하여 유출량을 예측하였으나 최근에는 인공신경망과 퍼지모형 그리고 유전자 알고리즘과 같은 인공지능기반의 모형들이 많이 사용되고 있다. 하지만 유출량을 예측하고자 할 때 학습자료 및 검정자료로써 사용되는 유출량은 수위-유량 관계곡선식으로부터 구하는 경우가 대부분으로 이렇게 유도된 유출량의 경우 오차가 크기 때문에 그 신뢰성에 문제가 있을 것으로 판단된다. 따라서 본 논문에서는 선행우량 및 수위자료로부터 단시간 수위예측에 관해 연구하였다. 신경망은 과거자료의 입 출력 패턴에서 정보를 추출하여 지식으로 보유하고, 이를 근거로 새로운 상황에 대한 해답을 제시하도록 하는 인공지능분야의 학습기법으로 인간이 과거의 경험과 훈련으로 지식을 축적하듯이 시스템의 입 출력에 의하여 연결강도를 최적화함으로서 모형의 구조를 스스로 조직화하기 때문에 모형의 구조에 적합한 최적 매개변수를 추정할 수 있다. 따라서 정확한 예측이 어려운 하천수위를 과거의 자료로 부터 학습된 신경망의 수학적 알고리즘을 통해 유출량의 예측에 적용할 수 있을 것이다. 유전자 알고리즘은 적자생존의 생물학 원리에 바탕을 둔 최적화 기법중의 하나로 자연계의 생명체 중 환경에 잘 적응한 개체가 좀 더 많은 자손을 남길 수 있다는 자연선택 과정과 유전자의 변화를 통해서 좋은 방향으로 발전해 나간다는 자연 진화의 과정인 자연계의 유전자 메커니즘에 바탕을 둔 탐색 알고리즘이다. 즉, 자연계의 유전과 진화 메커니즘을 공학적으로 모델화함으로써 잠재적인 해의 후보들을 모아 군집을 형성한 뒤 서로간의 교배 혹은 변이를 통해서 최적 해를 찾는 계산 모델이다. 따라서 본 연구에서는 인공신경망의 가중치를 유전자 알고리즘에 의해 최적화시킨후 오류역전파알고리즘에 의해 신경망의 학습을 진행하는 모형으로 감천유역의 선산수위표지점의 수위를 1시간~6시간까지 예측하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.2C
• /
• pp.31-40
• /
• 2005

we propose an enhanced self-generation supervised algorithm that by combining an ART algorithm and the delta-bar-delta method. Form the input layer to the hidden layer, ART-1 and ART-2 are used to produce nodes, respectively. A winner-take-all method is adopted to the connection weight adaption so that a stored pattern for some pattern is updated. we test the recognition of student identification, a certificate of residence, and an identifier from container that require nodes of hidden layers in neural network. In simulation results, the proposed self-generation supervised learning algorithm reduces the possibility of local minima and improves learning speed and paralysis than conventional neural networks.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1993.10a
• /
• pp.488-492
• /
• 1993

This paper proposes a new approach to the optimization method of a blending process with neural network. The method is based on the error backpropagation learning algorithm for neural network. Since the neural network can model an arbitrary nonlinear mapping, it is used as a system solver. A fuzzy membership function is used in parallel with the neural network to minimize the difference between measurement value and input value of neural network. As a result, we can guarantee the reliability and stability of blending process by the help of neural network and fuzzy membership function.

• Geotechnical Engineering
• /
• v.10 no.3
• /
• pp.79-96
• /
• 1994

A tunnel design expert system entitled NESTED is developed using the artificial neural network. The expert system includes three neural network computer models designed for the stability assessment of underground openings and the estimation of correlation between the RMR and Q systems. The expert system consists of the three models and the computerized rock mass classification programs that could be driven under the same user interface. As the structure of the neural network, a multi -layer neural network which adopts an or ror back-propagation learning algorithm is used. To set up its knowledge base from the prior case histories, an engineering database which can control the incomplete and erroneous information by learning process is developed. A series of experiments comparing the results of the neural network with the actual field observations have demonstrated the inferring capabilities of the neural network to identify the possible failure modes and the support timing. The neural network expert system thus complements the incomplete geological data and provides suitable support recommendations for preliminary design of tunnels in rock masses.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.1
• /
• pp.378-384
• /
• 2013

In this paper, we propose the fuzzy neural networks based on fuzzy c-means clustering algorithm. Typically, the generation of fuzzy rules have the problem that the number of fuzzy rules exponentially increases when the dimension increases. To solve this problem, the fuzzy rules of the proposed networks are generated by partitioning the input space in the scatter form using FCM clustering algorithm. The premise parameters of the fuzzy rules are determined by membership matrix by means of FCM clustering algorithm. The consequence part of the rules is expressed in the form of polynomial functions and the learning of fuzzy neural networks is realized by adjusting connections of the neurons, and it follows a back-propagation algorithm. The proposed networks are evaluated through the application to nonlinear process.

• Journal of Navigation and Port Research
• /
• v.32 no.8
• /
• pp.589-596
• /
• 2008

In this paper, Artificial Neural Network(ANN) is applied to automatic berthing control for a ship. ANN is suitable for a maneuvering such as ship's berthing, because it can describe non-linearity of the system. Multi-layer perceptron which has more than one hidden layer between input layer and output layer is applied to ANN. Using a back-propagation algorithm with teaching data, we trained ANN to get a minimal error between output value and desired one. For the automatic berthing control of a containership, we introduced low speed maneuvering mathematical models. The berthing control with the structure of 8 input layer units in ANN is compared to 6 input layer units. From the simulation results, the berthing conditions are satisfied, even though the berthing paths are different.

• The KIPS Transactions:PartB
• /
• v.9B no.1
• /
• pp.119-128
• /
• 2002

This paper proposes an efficient recognition method of car license plate from the car images by using both the dynamical thresholding and the neural network with enhanced learning algorithm. The car license plate is extracted by the dynamical thresholding based on the structural features and the density rates. Each characters and numbers from the p]ate is also extracted by the contour tracking algorithm. The enhanced neural network is proposed for recognizing them, which has the algorithm of combining the modified ART1 and the supervised learning method. The proposed method has applied to the real-world car images. The simulation results show that the proposed method has better the extraction rates than the methods with information of the gray brightness and the RGB, respectively. And the proposed method has better recognition performance than the conventional backpropagation neural network.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.8
• /
• pp.11-18
• /
• 2009

This paper proposes the way of improving learning speed in Levenberg-Marquardt algorithm using the principal submatrix of Jacobian matrix. The Levenberg-Marquardt learning uses Jacobian matrix for Hessian matrix to get the second derivative of an error function. To make the Jacobian matrix an invertible matrix. the Levenberg-Marquardt learning must increase or decrease ${\mu}$ and recalculate the inverse matrix of the Jacobian matrix due to these changes of ${\mu}$. Therefore, to have the proper ${\mu}$, we create the principal submatrix of Jacobian matrix and set the ${\mu}$ as the eigenvalues sum of the principal submatrix. which can make learning speed improve without calculating an additional inverse matrix. We also showed that our method was able to improve learning speed in both a generalized XOR problem and a handwritten digit recognition problem.

• Journal of KIISE:Software and Applications
• /
• v.31 no.9
• /
• pp.1171-1179
• /
• 2004

We describe word boundary detection that extracts the boundary between speech and non-speech. The proposed method uses two features. One is the normalized root mean square of speech signal, which is insensitive to white noises and represents temporal information. The other is the normalized met-frequency band energy of voice signal, which is frequency information of the signal. Our method detects word boundaries using a recurrent fuzzy associative memory(RFAM) that extends FAM by adding recurrent nodes. Hebbian learning method is employed to establish the degree of association between an input and output. An error back-propagation algorithm is used for teaming the weights between the consequent layer and the recurrent layer. To confirm the effectiveness, we applied the suggested system to voice data obtained from KAIST.