• 한국정보처리학회논문지
• /
• 제4권11호
• /
• pp.2780-2785
• /
• 1997

지식기반인공신경 망은 다른 기계학습알고리즘보다 우수한 성능을 나타내지만 인공신경망으로 형성된 후 동적으로 그 구조를 변경할 수 없어서 영역이론정련화 기능을 갖추지 못하였다. 지식기반인공신경망의 이러한 단점을 보완하기 위하여 TopGen 알고리즘이 제안되었으나 삽입된 은닉노드를 모든 입력 노드에 연결한 점, 빔탐색을 이용한 점 등의 문제를 안고 있다. 본 논문에서는 TopGen의 문제점을 해소하기 위하여 은닉노드를 입력 노드 중 관계가 깊은 일부의 노드에만 링크시켰으며, 역추적을 허용한 언덕오르기를 이용하는 알고리즘을 설계 하였다．

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2004년도 학술발표회
• /
• pp.1285-1289
• /
• 2004

Elman Discrete Recurrent Neural Networks Model(EDRNNM) was used to be a suitable short-term hydrological forecasting tool yielding a very high degree of flood stage forecasting accuracy at Musung station of Wi-stream one of IHP representative basins in South Korea. A relative new approach method has recurrent feedback nodes and virtual small memory in the structure. EDRNNM was trained by using two algorithms, namely, LMBP and RBP The model parameters, optimal connection weights and biases, were estimated during training procedure. They were applied to evaluate model validation. Sensitivity analysis test was also performed to account for the uncertainty of input nodes information. The sensitivity analysis approach could suggest a reduction of one from five initially chosen input nodes. Because the uncertainty of input nodes information always result in uncertainty in model results, it can help to reduce the uncertainty of EDRNNM application and management in small catchment.

• 전기학회논문지
• /
• 제66권1호
• /
• pp.137-143
• /
• 2017

In this paper, we investigate static control of Boolean networks described in the framework of semi-tensor product (STP) operation. The control objective is to determine control input nodes and their logical values so as to stabilize the considered Boolean network to a desired fixed point or cycle. Using topology of Boolean networks such as incidence matrix and hub nodes, a set of appropriate control input nodes is selected, and based on STP operations, we assign constant control inputs so that the controlled network can converge to a prescribed fixed point or cycle. To validate applicability of the proposed scheme, we conduct a numerical study on the problem of determining control input nodes for a Boolean network representing hierarchical differentiation of myeloid progenitors.

• 한국경영과학회지
• /
• 제31권3호
• /
• pp.41-51
• /
• 2006

Clustering is to group similar objects into clusters. Until now there are a lot of approaches using Self-Organizing feature Maps (SOFMS) But they have problems with a small output-layer nodes and initial weight. For example, one of them is a one-dimension map of c output-layer nodes, if they want to make c clusters. This approach has problems to classify elaboratively. This Paper suggests one-dimensional output-layer nodes in SOFMs. The number of output-layer nodes is more than those of clusters intended to find and the order of output-layer nodes is ascending in the sum of the output-layer node's weight. We un find input data in SOFMs output node and classify input data in output nodes using Euclidean distance. The proposed algorithm was tested on well-known IRIS data and TSPLIB. The results of this computational study demonstrate the superiority of the proposed algorithm.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(2)
• /
• pp.101-104
• /
• 2002

This paper propose that is algorithm of power dissipation reduction in the high level synthesis design for DSP(Digital Signal Processor), as the portable terminal system recently demand high power dissipation. This paper obtain effect of power dissipation reduction and switching activity that increase correlation of operands as input data of function unit. The algorithm search loop or repeatedly data to the input operands of function unit. That can be reduce the power dissipation using the new low power high level synthesis algorithm. In this Paper, scheduling operation search same nodes from input DFG(Data Flow Graph) with correlation coefficient of first input node and among nodes. Function units consist a multiplier, an adder and a register. The power estimation method is added switching activity for each bits of nodes. The power estimation have good efficient using proposed algorithm. This paper result obtain more Power reduction of fifty percents after using a new low power algorithm in a function unit as multiplier.

• 전기의세계
• /
• 제22권4호
• /
• pp.25-29
• /
• 1973

The computation of transmittances between arbitrary input and output nodes is of particular interest in the signal flow graph theory imput. The signal flow matrix [T] can be defined by [X]=-[T][X] where [X] and [Y] are input nose and output node matrices, respectively. In this paper, the followings are discussed; 1) Reduction of nodes by reforming the signal flow matrix., 2) Solution of input-output relationships by means of Gauss-Jordan reduction method, 3) Extension of the above method to the matrix signal flow graph.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.42.6-42
• /
• 2002

Kohonen's self organizing feature map (SOFM) converts arbitrary dimensional patterns into one or two dimensional arrays of nodes. Among the many competitive learning algorithms, SOFM proposed by Kohonen is considered to be powerful in the sense that it not only clusters the input pattern adaptively but also organize the output node topologically. SOFM is usually used for a preprocessor or cluster. It can perform dimensional reduction of input patterns and obtain a topology-preserving map that preserves neighborhood relations of the input patterns. The traditional SOFM algorithm[1] is a competitive learning neural network that maps inputs to discrete points that are called nodes on a lattice...

• 산업경영시스템학회지
• /
• 제23권61호
• /
• pp.127-135
• /
• 2000

This paper provides a new method of initializing neurons used in self-organizing neural networks and sequencing input nodes for applying to Euclidean traveling salesman problem. We use a general property that in any optimal solution for Euclidean traveling salesman problem, vertices located on the convex hull are visited in the order in which they appear on the convex hull boundary. We composite input nodes as number of convex hulls and initialize neurons as shape of the external convex hull. And then adapt input nodes as the convex hull unit and all convex hulls are adapted as same pattern, clockwise or counterclockwise. As a result of our experiments, we obtain l∼3 % improved solutions and these solutions can be used for initial solutions of any global search algorithms.

• 한국정밀공학회지
• /
• 제10권4호
• /
• pp.170-179
• /
• 1993

In this study, self-organized neural network is used to solve the vorrespondence problem of the axial stereo image. Edge points are extracted from a pair of stereo images and then the edge points of rear image are assined to the output nodes of neural network. In the matching process, the two input nodes of neural networks are supplied with the coordi- nates of the edge point selected randomly from the front image. This input data activate optimal output node and its neighbor nodes whose coordinates are thought to be correspondence point for the present input data, and then their weights are allowed to updated. After several iterations of updating, the weights whose coordinates represent rear edge point are converged to the coordinates of the correspondence points in the front image. Because of the feature map properties of self-organized neural network, noise-free and smoothed depth data can be achieved.

• 전자공학회논문지CI
• /
• 제47권5호
• /
• pp.25-34
• /
• 2010

스타이너 포인트들을 추가하여 모든 입력 노드들을 최단 길이로 연결하는 스타이너 최소 트리는 최소 신장 트리에 비해 전체 길이는 짧으나, 그것을 생성하는 문제는 NP-Complete 영역에 속한다. 이 문제를 위한 휴리스틱들은, 입력 노드의 수가 매우 큰 경우에는 많은 시간과 계산을 요구한다. 본 논문에서는 많은 입력 노드에 대해, 최하위 계층에서 포탈을 이용한 모든 가능한 단위 스타이너 트리들을 생성하고 각 상위 계층에서 이들을 계층별 병합 처리하여 최상위 계층에서 최소 비용의 트리를 선택하는 효과적인 PTAS 기법을 제안한다. 16,000개의 입력 노드와 최하위 계층에서 16개의 단위 영역으로 설계된 실험에서 생성된 PTAS 스타이너 트리는, pure 스타이너 트리의 길이에 비해 길이가 0.24% 증가되었으나, 생성 시간은 직렬 처리는 85.4%, 병렬처리는 98.9% 개선되었다. 따라서 제안하는 PTAS 스타이너 트리 생성 기법은 많은 입력 노드들에 대해 근사 스타이너 트리를 신속히 생성하는 응용에 잘 적용될 수 있을 것이다.