• Communications for Statistical Applications and Methods
• /
• v.17 no.6
• /
• pp.803-810
• /
• 2010

The singular value decomposition of a rectangular matrix is a basic tool to understand the structure of the data and particularly the relationship between row and column factors. However, conventional singular value decomposition used the least squares method and is not robust to outliers. We propose a simple robust singular value decomposition algorithm based on the weighted least absolute deviation which is not sensitive to leverage points. Its implementation is easy and the computation time is reasonably low. Numerical results give the data structure and the outlying information.

• Proceedings of the KIEE Conference
• /
• 1987.07a
• /
• pp.185-189
• /
• 1987

An analysis and design of large-scale linear multivariable systems often requires to be block triangularized form for good sensitivity of the systems when their poles and zeros are varied. But the decomposition algorithms presented up to now need a procedure of permutation, rescaling and a solution of nonlinear algebraic equations, which are usually burden. To avoid these problem, in this paper we develop a newly alternative block triangular decomposition algorithm which used the generalized matrix sign function on the Z-plane. Also, the decomposition algorithm demonstrated using the fifth order linear model of a distillation tower system.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.7
• /
• pp.1014-1018
• /
• 2012

This paper presents a decentralized controller design for formation flying of multiple satellites based on the overlapping decomposition technique. Each satellite is assumed to avail only the information of its own and in front of itself, which restricts the structure of a controller gain matrix to an overlapped form. The concerned large-scale system is expanded using the overlapping decomposition technique. Design condition is represented in terms of linear matrix inequalities with small-scale systems in a decentralized form, based on the expanded system. The resulting controller is contracted to the original overlapped form so as to close the original system. A numerical simulation shows the effectiveness of the proposed technique.

• Journal of Korea Multimedia Society
• /
• v.22 no.1
• /
• pp.35-43
• /
• 2019

Template tracking refers to the procedure of finding the most similar image patch corresponding to the given template through an image sequence. In order to obtain more accurate trajectory of the template, the template requires to be updated to reflect various appearance changes as it traverses through an image sequence. To do that, appearance images are used to model appearance variations and these are obtained by the computation of the principal components of the augmented image matrix at every iteration. Unfortunately, it is prohibitively expensive to compute the principal components at every iteration. Thus in this paper, we suggest a new Sliding Window based truncated URV Decomposition (TURVD) algorithm which enables updating their structure by recycling their previous decomposition instead of decomposing the image matrix from the beginning. Specifically, we show an efficient algorithm for updating and downdating the TURVD simultaneously, followed by the rank-one update to the TURVD while tracking the decomposition error accurately and adjusting the truncation level adaptively. Experiments show that the proposed algorithm produces no-meaningful differences but much faster execution speed compared to the typical algorithms in template tracking applications, thereby maintaining a good approximation for the principal components.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.3
• /
• pp.163-168
• /
• 2011

In this paper, we proposed non-linear gamma curve algorithm for gamma correction. The previous non-linear gamma curve algorithm is generated by the least square polynomial using the Gauss-Jordan inverse matrix. However, the previous algorithm has some weak points. When calculating coefficients using inverse matrix of higher degree, occurred truncation errors. Also, only if input sample points are existed regular interval on 10-bit scale, the least square polynomial is accurately works. To compensate weak-points, we calculated accurate coefficients of polynomial using eigenvalue and orthogonal value of mat11x from singular value decomposition (SVD) and QR decomposition of vandemond matrix. Also, we used input data part segmentation, then we performed polynomial curve fitting and merged curve fitting results. When compared the previous method and proposed method using the mean square error (MSE) and the standard deviation (STD), the proposed segmented polynomial curve fitting is highly accuracy that MSE under the least significant bit (LSB) error range is approximately $10^{-9}$ and STD is about $10^{-5}$.

• Architectural research
• /
• v.14 no.4
• /
• pp.143-152
• /
• 2012

A form-finding procedure is presented for planar tensegrity structures. Notably, a simple decision criteria is proposed to select the desirable candidate position vector from the unitary matrix produced by the eigenvalue decomposition of force density matrix. The soundness of the candidate position vector guarantees faster convergence and produces a desirable form of tensegrity without any member having zero-length. Several numerical examples are provided to demonstrate the capability of the proposed form-finding process.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.2
• /
• pp.288-294
• /
• 2008

Motivated by the fact that upper solution bounds for the continuous Lyapunov equation are valid under some very restrictive conditions, an attempt is made to extend the set of Hurwitz matrices for which such bounds are applicable. It is shown that the matrix set for which solution bounds are available is only a subset of another stable matrices set. This helps to loosen the validity restriction. The new bounds are illustrated by examples.

• The Pure and Applied Mathematics
• /
• v.2 no.2
• /
• pp.133-140
• /
• 1995

The D-optimal design criterion for precise parameter estimation in nonlinear regression analysis is called the determinant criterion because the determinant of a matrix is to be maximized. In this thesis, we derive the gradient and the Hessian of the determinant criterion, and apply a QR decomposition for their efficient computations. We also propose an approximate form of the Hessian matrix which can be calculated from the first derivative of a model function with respect to the design variables. These equations can be used in a Gauss-Newton type iteration procedure.

• The KIPS Transactions:PartB
• /
• v.13B no.6 s.109
• /
• pp.625-632
• /
• 2006

Collaborative filtering is a technology that aims at teaming predictive models of user preferences. Collaborative filtering systems have succeeded in Ecommerce market but they have shortcomings of high dimensionality and sparsity. In this paper we propose the nearest neighbor collaborative filtering method using non-negative matrix factorization(NNMF). We replace the missing values in the user-item matrix by using the user variance coefficient method as preprocessing for matrix decomposition and apply non-negative factorization to the matrix. The positive decomposition method using the non-negative decomposition represents users as semantic vectors and classifies the users into groups based on semantic relations. We compute the similarity between users by using vector similarity and selects the nearest neighbors based on the similarity. We predict the missing values of items that didn't rate by a new user based on the values that the nearest neighbors rated items.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.3
• /
• pp.815-825
• /
• 2016

Marginalized random effects models (MREMs) are often used to analyze longitudinal categorical data. The models permit direct estimation of marginal mean parameters and specify the serial correlation of longitudinal categorical data via the random effects. However, it is not easy to estimate the random effects covariance matrix in the MREMs because the matrix is high-dimensional and must be positive-definite. To solve these restrictions, we introduce two modeling approaches of the random effects covariance matrix: partial autocorrelation and the modified Cholesky decomposition. These proposed methods are illustrated with the real data from Korean genomic epidemiology study.