• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1245-1255
• /
• 2017

The online learning is a process of obtaining the solution for a given objective function where the data is accumulated in real time or in batch units. The stochastic gradient descent method is one of the most widely used for the online learning. This method is not only easy to implement, but also has good properties of the solution under the assumption that the generating model of data is homogeneous. However, the stochastic gradient method could severely mislead the online-learning when the homogeneity is actually violated. We assume that there are two heterogeneous generating models in the observation, and propose the a new stochastic gradient method that mitigate the problem of the heterogeneous models. We introduce a robust mini-batch optimization method using statistical tests and investigate the convergence radius of the solution in the proposed method. Moreover, the theoretical results are confirmed by the numerical simulations.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.925-928
• /
• 2008

With basis of the development of information communication techniques in recent year, the digital processing techniquy also has been growed fast. The digital processing technique have used signals - speech and image processing- for processing of transmission and analysis. After we get and save the signals. Effective signal processing techniques have varied filters and typical digital filters are FIR filter and IIR filter. The FIR digital filter is more secure because phase response characteristics have linear phase. But, FIR digital filters have a problem to product the Gibbs phenomenon generating around a discontinuous point. A propose of filer is to remove the problem. Therefore, in this paper I was proposed a method using FIR digital filter applied a modified window function and the method was compared with conventional methods.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.5
• /
• pp.1289-1296
• /
• 1997

To build up a knowledge-based system in an Artifical Inerligence System, selecting an appropriate set of rules is one of the key provlems. In this paper, we discuss a new method for exteacting fuzzy rules diredtly from fuzzy membdrchip function dat for pattern classifcation. The fuzzy rules with variable fuzzy recions are defined by sharing fuzzy space in fuzzy grid.Tehse rules are extracted form memberchop function. Them, optimal input vari-ables for the rules are determined using the number of extracted rules as a criterion. The method is compared with neural networks using Ishibuchi. Finally, in order to demonstrate the cffectiveness of the present method, simulation results are shown.

• The Korean Journal of Applied Statistics
• /
• v.23 no.3
• /
• pp.521-531
• /
• 2010

Interval-censored observations are common in medical and epidemiologic studies; however, limited studies exist due to the complexity and special structure of interval-censoring. This paper introduces the imputation method and the self consistency method in the interval-censored data. We propose a new method of generating random numbers under an interval-censoring set-up. Through simulation studies we compare two methods under various simulation schemes in the sense of the mean squared error for estimating the median survival time and the mean integrated squared error for estimating the survival function. Under a moderate censoring percentage, the mean imputation method showed a better performance than the self-consistency method in estimating the median survival time and the survival function.

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.2
• /
• pp.76-84
• /
• 2017

This paper presents an efficient method for generating spatiotemporal images in order to detect leukocyte motion in microvessels. Leveraging the constraint that leukocytes move along the contour line of the blood vessel wall, our proposed method efficiently generates spatiotemporal images for leukocyte motion detection. To that end, translational motion caused by in vivo movement is first removed by a template matching method. Second, the blood vessel region is detected by an automatic threshold selection method in order to binarize temporal variance images. Then, the contour of the blood vessel wall is expressed via B-spline function. Finally, using the detected blood vessel wall's contour as an initial curve, the plasma layer for the most accurate position is determined in order to find the spatial axis via snake, and the spatiotemporal images are generated. Experimental results show that the spatiotemporal images are generated effectively through comparison of each step with three images.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2007.05a
• /
• pp.928-933
• /
• 2007

In this research, we investigated active vibration control based on wave-viewpoint using the longitudinal wave separation method. The control strategy is the one of active vibration control technique for generating vibration reduced zone and uses wave information including the directivity as the cost function. In order to get the wave information from the measured values, we proposed and examined the time-domain longitudinal wave separation method proper to real time application like active vibration control. Using the proposed method, we examine the performance and feasibility of active vibration control based wave view-point through the simulation. The related experimental verification and application is going to be expected in a near future.

• Structural Engineering and Mechanics
• /
• v.30 no.1
• /
• pp.119-129
• /
• 2008

A small strain and elastoplastic formulation of Polygonal Element Method (PEM) is developed for efficient analysis of elastoplastic solids. In this work, the polygonal elements are constructed based on traditional triangular finite meshes. The construction method of polygonal mesh can directly utilize the sophisticated triangularization algorithm and reduce the difficulty in generating polygonal elements. The Wachspress rational finite element basis function is used to construct the approximations of polygonal elements. The incremental variational form and a von Mises type model are used for non-linear elastoplastic analysis. Several small strain elastoplastic numerical examples are presented to verify the advantages and the accuracy of the numerical formulation.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2001.10a
• /
• pp.220-226
• /
• 2001

It is important to estimate exactly wave forces acting on various types of offshore structures under the severe environmental conditions in the ocean site. This paper presents an easy experimental method which deals with transient waves. The proposed scheme made it possible to generate breaking waves at any position in the wave tank by changing the maximum slope of the component waves. The theoretical and experimental methods were investigated by generating concentrated waves which acted on a single and multiple cylinders. The waves forces increased rapidly when the models encountered breaking waves. The theoretical results underestimates the forces due to breaking waves. Therefore, the effects due to breaking waves should be considered carefully in the design process of a structure under the influence of breaking waves.

• Journal of Ocean Engineering and Technology
• /
• v.15 no.4
• /
• pp.8-13
• /
• 2001

It is important to estimate exactly wave forces acting on various types of offshore structures under the severe environmental conditions in the ocean site. This paper presents an easy experimental method which deals with transient waves. The proposed scheme made it possible to generate breaking waves at any position in the wave tank by changing the maximum slope of the component waves. The theoretical and experimental methods were investigated by generating concentrated waves which acted on a single and multiple cylinders. The waves forces increased rapidly when the models encountered breaking waves. The theoretical results underestimates the forces due to breaking waves. Therefore, the effects due to breaking waves should be considered carefully in the design process of a structure under the influence of breaking waves.

• Journal of Ship and Ocean Technology
• /
• v.11 no.4
• /
• pp.21-28
• /
• 2007

Recently the importance of the utilization of engineering data is gradually increasing. Engineering data contains the experiences and know-how of experts. Data mining technique is useful to extract knowledge or information from the accumulated existing data. This paper deals with generating optimal polynomials using genetic programming (GP) as the module of Data Mining system. Low order Taylor series are used to approximate the polynomial easily as a nonlinear function to fit the accumulated data. The overfitting problem is unavoidable because in real applications, the size of learning samples is minimal. This problem can be handled with the extended data set and function node stabilization method. The Data Mining system for the ship design based on polynomial genetic programming is presented.