• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.60-64
• /
• 2001

In this paper, a new learning methodology for Kernel Methods is suggested that results in a sparse representation of kernel space from the training patterns for classification problems. Among the traditional algorithms of linear discriminant function(perceptron, relaxation, LMS(least mean squared), pseudoinverse), this paper shows that the relaxation procedure can obtain the maximum margin separating hyperplane of linearly separable pattern classification problem as SVM(Support Vector Machine) classifier does. The original relaxation method gives only the necessary condition of SV patterns. We suggest the sufficient condition to identify the SV patterns in the learning epochs. Experiment results show the new methods have the higher or equivalent performance compared to the conventional approach.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.9
• /
• pp.817-821
• /
• 2001

In this paper, a new learning methodology for kernel methods that results in a sparse representation of kernel space from the training patterns for classification problems is suggested. Among the traditional algorithms of linear discriminant function, this paper shows that the relaxation procedure can obtain the maximum margin separating hyperplane of linearly separable pattern classification problem as SVM(Support Vector Machine) classifier does. The original relaxation method gives only the necessary condition of SV patterns. We suggest the sufficient condition to identify the SV patterns in the learning epoches. For sequential learning of kernel methods, extended SVM and kernel discriminant function are defined. Systematic derivation of learning algorithm is introduced. Experiment results show the new methods have the higher or equivalent performance compared to the conventional approach.

• Structural Engineering and Mechanics
• /
• v.18 no.5
• /
• pp.565-589
• /
• 2004

The present paper concerns the macroscopic overall description of rheologic properties for steel wire and synthetic fibre cables under variable loading actions according to non-linear creep and/or relaxation theory. The general constitutive equations of non-linear creep and/or relaxation of tension elements - cables under one-step and the variable stress or strain inputs using the product and two types of additive approximations of the kernel functions are presented in the paper. The derived non-linear constitutive equations describe a non-linear rheologic behaviour of the cables for a variable stress or strain history using the kernel functions determined only by one-step - constant creep or relaxation tests. The developed constitutive equations enable to simulate and to predict in a general way non-linear rheologic behaviour of the cables under an arbitrary loading or straining history. The derived constitutive equations can be used for the various tension structural elements with the non-linear rheologic properties under uniaxial variable stressing or straining.

• The KIPS Transactions:PartB
• /
• v.9B no.6
• /
• pp.735-744
• /
• 2002

This paper proposes learning performance improvement of support vector machine using the kernel relaxation and the dynamic momentum. The dynamic momentum is reflected to different momentum according to current state. While static momentum is equally influenced on the whole, the proposed dynamic momentum algorithm can control to the convergence rate and performance according to the change of the dynamic momentum by training. The proposed algorithm has been applied to the kernel relaxation as the new sequential learning method of support vector machine presented recently. The proposed algorithm has been applied to the SONAR data which is used to the standard classification problems for evaluating neural network. The simulation results of proposed algorithm have better the convergence rate and performance than those using kernel relaxation and static momentum, respectively.

• Journal of Korea Multimedia Society
• /
• v.6 no.5
• /
• pp.788-796
• /
• 2003

This paper proposes using dynamic momentum for squential learning method. Using The dynamic momentum improves convergence speed and performance by the variable momentum, also can identify it in the RMSE(root mean squared error). The proposed method is reflected using variable momentum according to current state. While static momentum is equally influenced on the whole, dynamic momentum algorithm can control the convergence rate and performance. According to the variable change of momentum by training. Unlike former classification and regression problems, this paper confirms both performance and regression rate of the dynamic momentum. Using RMSE(root mean square error ), which is one of the regression methods. The proposed dynamic momentum has been applied to the kernel adatron and kernel relaxation as the new sequential learning method of support vector machine presented recently. In order to show the efficiency of the proposed algorithm, SONAR data, the neural network classifier standard evaluation data, are used. The simulation result using the dynamic momentum has a better convergence rate, performance and RMSE than those using the static moment, respectively.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.313-315
• /
• 2002

본 논문에서는 학습 중 모멘트를 동적으로 조절하여 수련속도와 학습 성능을 향상시키는 동적모멘트를 제안하고 회귀방법으로 동적모멘트의 성능을 재확인한다. 제안된 학습방법은 기존의 정적모멘트와는 달리 수렴 정도에 따라 현재의 학습에 과거의 학습률을 단리 반영하는 방법으로 다른 학습법에 비해 보다 유연한 초평면을 갖으며 수렴에 이르는 시간이 오래 걸리는 KR(Kernel Relaxation)에 적용하여 그 성능을 확인한다. 본 논문에서 사용한 회귀방법은 RMS 오류율을 사용하였으며 제안된 학습방법인 동적모멘트를 SVM(support vector machine)의 순차 학습방법 중 최근 발표된 KR에 적용하여 RMS 오류율을 확인하였다. 실험의 공정성을 위해 신경망 분류기 표준평가데이터인 SONAR 데이터를 사용하였으며 실험 결과 동적모멘트를 이용한 회귀율이 정적모멘트를 이용한 방법보다 향상되었음을 확인하였다.

• Journal of Biosystems Engineering
• /
• v.15 no.3
• /
• pp.207-218
• /
• 1990

Grains display characteristics of both elastic bodies and viscous fluids when they are subjected to mechanical treatments in harvesting, handling, and processing. This viscoelastic behavior of grains when mechanically stressed must be fully understood to establish maximum machine efficiency and have a minimum degree of grain damage and the highest quality of the final product. The studies were conducted to examine the effect of the moisture content, the loading rate and the initial deformation on the stress relaxation behavior of whole kernel of rough rice, and develop the rheological model to represent its stress relaxation behavior. The following results were obtained from the study. 1. Moisture content had the greatest influence on the initial portion of the relaxation curve. With elapsing time the lower moisture content resulted in the lower residual stress for the Japonica-type rough rice and vice versa for the Indica-type rough rice. But within the ranges of moisture content tested, the degree of stress relaxation per unit strain on the Indica-type rough rice was a little higher than those on the Japonica-type rough rice. 2. The slower loading rate resulted in less initial stress. The decreasing trend of residual stress for all the samples tested with increasing loading rate was shown. 3. The higher initial deformation for all the samples resulted in less initial stress. The increasing of amount of stress relaxation per unit strain with increase of initial stress indicated that viscoelastic properties of rough rice depended not only upon duration of load applied but also initial stress applied. This means that rough rice is nonlinear viscoelastic material. 4. The compression stress relaxation properties of rough rice kernel can be described by a generalized Maxwell model representing by the Maxwell elements.

• Proceedings of the IEEK Conference
• /
• 2002.06c
• /
• pp.9-12
• /
• 2002

This paper proposes a dynamic moment algorithm to control oscillaion before the convergence of the KR(Kernel Relaxation). The proposed dynamic moment algorithm can be controlled to convergence speed and performance according to the change of the dynamic moment by teaming training. we used SONAR data that is a neural network classifier standard evaluation data in order to do impartial performance evaluation. The proposed algorithm has been applied to the KP (kernel perceptron), KPM(kernel perceptron with margin) and KLMS(kernel lms) as the kernel method presented recently. The simulation results of proposed algorithm have better the convergence performance than those using none and static moment.

• Bulletin of the Korean Chemical Society
• /
• v.26 no.12
• /
• pp.1986-1990
• /
• 2005

We study the relaxation kinetics of reversible reactions of the type A + B $^\leftarrow_\rightarrow$ C + B by applying the manyparticle kernel theory, which we have developed to investigate many particle effects on general diffusioninfluenced reactions. It is shown that for the target model, where A and C molecules are immobile and their interconversion is induced by the encounter with the B molecules that are present in much excess, the manyparticle kernel theory gives a result that coincides with the known exact result.

• Journal of Biosystems Engineering
• /
• v.16 no.3
• /
• pp.263-271
• /
• 1991

When grains is subjected to oscillating load, the dynamic viscoelastic behavior of the material will be describe the complex modulus of the material. The complex modulus and therefore the storage modulus, the loss modulus, and the phase angle for the sample should be obtainable with a given static viscoelastic property of the material under static load. The complex relaxation moduli of the rough rice kernel were computed from the Burger's model describing creep behavior of the material which were obtained in the previous study. Also, the effects of cyclic load and moisture content of grain on the dynamic viscoelastic behavior of the samples were analized. The storage modulus of the rough rice kernel slightly increased with the frequency applied but at above the frequency of 0.1 Hz it was nearly constant with the frequency, and the loss modulus of the sample very rapidly decreased with increase in the frequency on those frequency ranges. It was shown that the storage modulus and the loss modulus of the sample increased with decrease in grain moisture content. Effect of grain moisture content on the storage modulus of the sample was highly significant than effect of the frequency applied, but effect of the frequency on the loss modulus of the sample was more significant than effect of grain moisture content.