• 한국데이터정보과학회:학술대회논문집
• /
• 2004.10a
• /
• pp.53-59
• /
• 2004

Support vector machine(SVM) approach to regression can be found in information science literature. SVM implements the regularization technique which has been introduced as a way of controlling the smoothness properties of regression function. In this paper, we propose a new estimation method based on quadratic loss SVM for a linear fuzzy regression model of Tanaka's, and furthermore propose a estimation method for nonlinear fuzzy regression. This approach is a very attractive approach to evaluate nonlinear fuzzy model with crisp input and output data.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.597-602
• /
• 2008

Most of structural analyses are concerned with the deformation and stress in a body subjected to external loads. In many fields, however, the interpretation of inverse problems is needed to determine surface tractions or internal stresses. In this study, the inverse processes by using the finite elements and the boundary elements are formulated for the evaluation of internal residual stresses from displacements measured on a remote surface. Small errors in the measured displacements often result in a substantial loss of accuracy of an inverse system. We use the Tikhonov regularization techniques to regularize the ill-conditioned system. Advantages and disadvantages are discussed through numerical examples.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.1
• /
• pp.83-85
• /
• 2016

The OFDM with LS signal detection performs worse in fast time varying channels as the channel matrix has higher chance of becoming ill-conditioned. Various regularization methods are applied to avoid performance degradation in LS signal detection. In this paper, we proposed a CGLS method with the stopping criteria imposed by the characteristics of the modulation method, which shows performance comparable to that of the optimal LMMSE.

• Communications for Statistical Applications and Methods
• /
• v.24 no.2
• /
• pp.143-154
• /
• 2017

A variable selection method based on probabilistic principal component analysis (PCA) using penalized likelihood method is proposed. The proposed method is a two-step variable reduction method. The first step is based on the probabilistic principal component idea to identify principle components. The penalty function is used to identify important variables in each component. We then build a model on the original data space instead of building on the rotated data space through latent variables (principal components) because the proposed method achieves the goal of dimension reduction through identifying important observed variables. Consequently, the proposed method is of more practical use. The proposed estimators perform as the oracle procedure and are root-n consistent with a proper choice of regularization parameters. The proposed method can be successfully applied to high-dimensional PCA problems with a relatively large portion of irrelevant variables included in the data set. It is straightforward to extend our likelihood method in handling problems with missing observations using EM algorithms. Further, it could be effectively applied in cases where some data vectors exhibit one or more missing values at random.

• Structural Engineering and Mechanics
• /
• v.54 no.6
• /
• pp.1217-1244
• /
• 2015

In this research a theoretical and numerical study on a bridge damage detection procedure is presented based on vibration measurements collected from a set of accelerometers. This method, referred to as "Adjoint Variable Method", is a sensitivity-based finite element model updating method. The approach relies on minimizing a penalty function, which usually consists of the errors between the measured quantities and the corresponding predictions attained from the model. Moving mass is an interactive model and includes inertia effects between the model and mass. This interactive model is a time varying system and the proposed method is capable of detecting damage in this variable system. Robustness of the proposed method is illustrated by correct detection of the location and extension of predetermined single, multiple and random damages in all ranges of speed and mass ratio of moving vehicle. A comparative study on common sensitivity and the proposed method confirms its efficiency and performance improvement in sensitivity-based damage detection methods. In addition various possible sources of error, including the effects of measurement noise and initial assumption error in stability of method are also discussed.

• Journal of Astronomy and Space Sciences
• /
• v.2 no.1
• /
• pp.19-33
• /
• 1985

The singularities of differential Newtonian equation of motion in three body problem cause the loss of accuracy and the considerable increase of the computer time. These singularities could be eliminated during the process of regularization to transform the independent variables and the coordinate of Newtonian equations of motion. In this study, we calculated the positions and velocities of three body along the time scale to find out the unique solution of regularized Newtonian equations of motion with the $5^{th}$ order Runge-Kutta method by assuming the suitable initial velocities and positions. As the results of these calculations it is shown that the tripe stellar system eventually distintegrated, two of them formed a binary, and the last one escaped from this system with a hyperbolic orbit. This may suggest one possible explanation for the binary formation.

• Journal of the Korea Concrete Institute
• /
• v.17 no.3 s.87
• /
• pp.369-374
• /
• 2005

This study proposes a technique for the recognition of crack patterns, which includes horizontal, vertical, diagonal($-45^{\circ}$), diagonal($+45^{\circ}$), and random cracks, based on image processing technique and artificial neural network. A MATLAB code was developed for the proposed image processing algorithm and artificial neural network. Features were determined using total projection technique, and the structure(no. of layers and hidden neurons) and weight of artificial neural network were determined by learning from artificial crack images. In this process, we adopted Bayesian regularization technique as a generalization method to eliminate overfitting Problem. Numerical tests were performed on thirty-eight crack images to examine validity of the algorithm. Within the limited tests in the present study, the proposed algorithm was revealed as accurately recognizing the crack patterns when compared to those classified by a human expert.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.25 no.6_1
• /
• pp.545-555
• /
• 2007

3D buildings are the most prominent feature comprising urban scene. A few of mega-cities in the globe are virtually reconstructed in photo-realistic 3D models, which becomes accessible by the public through the state-of-the-art online mapping services. A lot of research efforts have been made to develop automatic reconstruction technique of large-scale 3D building models from remotely sensed data. However, existing methods still produce irregular building polygons due to errors induced partly by uncalibrated sensor system, scene complexity and partly inappropriate sensor resolution to observed object scales. Thus, a geometric regularization technique is urgently required to rectify such irregular building polygons that are quickly captured from low sensory data. This paper aims to develop a new method for regularizing noise building outlines extracted from airborne LiDAR data, and to evaluate its performance in comparison with existing methods. These include Douglas-Peucker's polyline simplication, total least-squared adjustment, model hypothesis-verification, and rule-based rectification. Based on Minimum Description Length (MDL) principal, a new objective function, Geometric Minimum Description Length (GMDL), to regularize geometric noises is introduced to enhance the repetition of identical line directionality, regular angle transition and to minimize the number of vertices used. After generating hypothetical regularized models, a global optimum of the geometric regularity is achieved by verifying the entire solution space. A comparative evaluation of the proposed geometric regulator is conducted using both simulated and real building vectors with various levels of noise. The results show that the GMDL outperforms the selected existing algorithms at the most of noise levels.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.4
• /
• pp.223-231
• /
• 2019

A continuum-based design sensitivity analysis(DSA) method is developed for electrostatic problems. To consider high order objective functions, we use 9-node finite element basis functions for analysis and DSA methods. As the design variables are parameterized with B-spline functions, smooth boundary variations are naturally obtained. To solve mesh entanglement problems during the optimization process, a mesh regularization scheme is employed. By minimizing the Dirichlet energy functional, mesh uniformity can be automatically achieved. In numerical examples for maximizing dielectrophoresis forces, the numerical results are compared with well-known electrode geometries and the obtained characteristics are discussed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.15-23
• /
• 2022

Among the learning methods for Continuous Learning environments, "Learning without Forgetting" has fixed regularization strengths, which can lead to poor performance in environments where various data are received. We suggest a way to set weights variable by identifying the features of the data we want to learn. We applied weights adaptively using correlation and complexity. Scenarios with various data are used for evaluation and experiments showed accuracy increases by up to 5% in the new task and up to 11% in the previous task. In addition, it was found that the adaptive weight value obtained by the algorithm proposed in this paper, approached the optimal weight value calculated manually by repeated experiments for each experimental scenario. The correlation coefficient value is 0.739, and overall average task accuracy increased. It can be seen that the method of this paper sets an appropriate lambda value every time a new task is learned, and derives the optimal result value in various scenarios.