• Journal of Korean Society of Transportation
• /
• v.18 no.5
• /
• pp.99-107
• /
• 2000

Path-based assignment(PBA) is valuable to dynamic traffic control and routing in integrated ITS framework. As one of widely studied PBA a1gorithms, Gradient Projection(GP) a1gorithm typically fields rapid convergence to a neighborhood of an optimal solution. But once it comes near a solution, it tends to slow down. To overcome this problem, we develop more efficient path-based assignment algorithm by combining Conjugate Gradient method with GP algorithm. It determines more accurate moving direction near a solution in order to gain a significant advantage in speed of convergence. Also this algorithm is applied to the Sioux-Falls network and verified its efficiency. Then we demonstrate that this type of method is very useful in improving speed of convergence in the case of user equilibrium problem.

• Structural Engineering and Mechanics
• /
• v.75 no.3
• /
• pp.339-356
• /
• 2020

In this article, a developed bond-based peridynamic model for functionally graded materials (FGMs) is proposed to simulate the dynamic fracture behaviors in FGMs. In the developed bond-based peridynamic model for FGMs, bonds are categorized into three different types, including transverse directionally peridynamic bond, gradient directionally peridynamic bond and arbitrary directionally peridynamic bond, according to the geometrical relationship between directions of peridynamic bonds and gradient bonds in FGMs. The peridynamic micromodulus in the gradient directionally and arbitrary directionally peridynamic bonds can be determined using the weighted projection method. Firstly, the standard bond-based peridynamic simulations of crack propagation and branching in the homogeneous PMMA plate are performed for validations, and the results are in good agreement with the previous experimental observations and the previous phase-field numerical results. Then, the numerical study of crack initiation, propagation and branching in FGMs are conducted using the developed bond-based peridynamic model, and the influence of gradient direction on the dynamic fracture behaviors, such as crack patterns and crack tip propagation speed, in FGMs is systematically studied. Finally, numerical results reveal that crack branching in FGMs under dynamic loading conditions is easier to occur as the gradient angle decreases, which is measured by the gradient direction and direction of the initial crack.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.823-826
• /
• 2000

Motion detection is an important step in computer vision and image processing. Traditional motion detection systems are classified into two categories, namely, feature based and gradient based. In feature based motion detection, features in consecutive frames are detected and matched. Gradient based methods assume that the intensity varies linearly and locally. The method, which we propose, is neither feature nor gradient based but uses the electric field theory. The pixels in an image are modeled as point charges and motion is detected by using the variations between the two electric fields produced by the charges corresponding to the two images.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.2-4
• /
• 2021

In a distributed machine learning system, sharing gradients was considered safe because it did not share original training data. However, recent studies found that malicious attacker could completely restore the original training data from shared gradients. Gradient Leakage Attack is a technique that restoring original training data by exploiting theses vulnerability. In this study, we present the image transformation method based on Discrete Cosine Transform to defend against the Gradient Leakage Attack on the federated learning setting, which training in local devices and sharing gradients to the server. Experiment shows that our image transformation method cannot be completely restored the original data from Gradient Leakage Attack.

• Multiscale and Multiphysics Mechanics
• /
• v.1 no.1
• /
• pp.15-33
• /
• 2016

In this paper, an asymptotic method is employed to formulate nano- or micro-beams based on strain gradient elasticity. Although a basic theory for the strain gradient elasticity has been well established in literature, a systematic approach is relatively rare because of its complexity and ambiguity of higher-order elasticity coefficients. In order to systematically identify the strain gradient effect, an asymptotic approach is adopted by introducing the small parameter which represents the beam geometric slenderness and/or the internal atomistic characteristic. The approach allows us to systematically split the two-dimensional strain gradient elasticity into the microscopic one-dimensional through-the-thickness analysis and the macroscopic one-dimensional beam analysis. The first-order beam problem turns out to be different from the classical elasticity in terms of the bending stiffness, which comes from the through-the-thickness strain gradient effect. This subsequently affects the second-order transverse shear stress in which the surface shear stress exists. It is demonstrated that a careful derivation of a first strain gradient elasticity embraces "Gurtin-Murdoch traction" as the surface effect of a one-dimensional Euler-Bernoulli-like beam model.

• Journal of the Korean Mathematical Society
• /
• v.56 no.4
• /
• pp.1001-1016
• /
• 2019

We are concerned with optimization methods for the $L^2$-Wasserstein least squares problem of Gaussian measures (alternatively the n-coupling problem). Based on its equivalent form on the convex cone of positive definite matrices of fixed size and the strict convexity of the variance function, we are able to present an implementable (accelerated) gradient method for finding the unique minimizer. Its global convergence rate analysis is provided according to the derived upper bound of Lipschitz constants of the gradient function.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.283-286
• /
• 2001

Based on a fuzzy system representation of gray scale images, we derive an edge detection algorithm whose convolution kernel is different from the known kernels such as those of Roberts', Prewitt's or Sobel's gradient. Our fuzzy system representation is an exact representation of the bicubic spline function which represents the gray scale image approximately. Hence the fuzzy system is a continuous function and it provides a natural way to define the gradient and the Laplacian operator. We show that the gradient at grid points can be evaluated by taking the convolution of the image with a 3 3 kernel. We also show that our gradient coupled with the approximate value of the continuous function generates an edge detection method which creates edge images clearer than those by other methods. A few examples of applying our methods are included.

• Proceedings of the KOR-KST Conference
• /
• 1999.10a
• /
• pp.65-70
• /
• 1999

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.9C
• /
• pp.882-888
• /
• 2006

Multiuser detection can significantly increase system capacity and improve service quality compared with the existing matched filter. In this paper, we introduce an method which efficiently calculates the maximum likelihood (ML) metric based on the gradient search (GS). The ML detection needs user powers as well as their spreading codes. A method is also proposed that allows us to detect data bits with the estimation of user powers when they are unknown. Computer simulation shows that the proposed method can nearly achieve the same performance as the GS with perfectly hewn user powers.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.52 no.9
• /
• pp.550-556
• /
• 2003

Price-based unit commitment is one of bidding strategies which a Genco may take in a practical manner. For that purpose, it is required for a Genco to decide output levels of its generators at each trade period. In this paper, an economic dispatch of thermal units is proposed considering the quantity of reserve contracts. A gradient projection algorithm is adopted as an optimization tool. A direct form of a projection matrix without any calculation of matrix inverse and multiplications is induced. Besides, it is proved that there is no need to check one of the two optimality conditions in the gradient projection method, which also requires matrix inverse and multiplications.