• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.484-484
• /
• 2000

The problem of determining the proper size of an neural network is recognized to be crucial, especially for its practical implications in such important issues as learning and generalization. Unfortunately, it usually is not obvious what size is best: a system that is too snail will not be able to learn the data while one that is just big enough may learn the slowly and be very sensitive to initial conditions and learning parameters. One popular technique is commonly known as pruning and consists of training a larger than necessary network and then removing unnecessary weights/nodes. In this paper, a new pruning method is developed, based on the penalty-term methods. This method makes the neural network good for the generalization and reduces the retraining time after pruning weights/nodes.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1997년도 가을 학술발표회 논문집
• /
• pp.163-170
• /
• 1997

The present paper describes an investigation into the application of the genetic algorithm(GA) in the optimization of structural design. Stochastic processes generate an initial population of designs and then apply principles of natural selection/survival of the fittest to improve the designs. The five test functions are used to verify the robustness and reliability of GA, and as a numerical example, minimum weight of a cantilever composite laminated beam with a mix of continuous, integer and discrete design variables is obtained by using GA with exterior penalty function method. The design problem has constraints on strength, displacements, and natural frequencies, and is formulated to a multidimensional nonlinear form. From the results, it is found that the GA search technique is very effective at finding the good optimum solution as well as has higher robustness.

• 한국경영과학회지
• /
• 제25권2호
• /
• pp.87-100
• /
• 2000

As a result of rapid advances in communication technology, fiber optics have begun to be adopted in most telecommunication systems 3s an economic choice Due to the trend of evolution toward broadband communication network with fiber optics and electronic devices. the network design problem for broadband communication has been received a great deal of research attention recently. In this paper, we address a topological design problem for broadband local access network with uncertain demands, which has received surprisingly little attention so far. in our problem, we select a set of hubs and links for constructing network expected penalty cost for the amount of undersupplied In addition to the usual cost terms of the fixed demand problem Our problem can be approximated as a mixed 0-1 integer programming problem by using Szwarc’s linear approximation technique. Then the problem is transformed successfully into a version of classical network design model. Some computational experiments for the model and concluding remarks are described.

• 한국통신학회논문지
• /
• 제32권5C호
• /
• pp.530-535
• /
• 2007

변이공간영상(Disparity space image, DSI) 방법은 스테레오 영상간의 정합을 찾는 방법이다. 이 방법은 각 화소 단위로 변이값을 계산해 내는 장점이 있다. DSI 방법은 비용함수를 최소화시키는 정합을 찾는 방법이다. 이 비용함수에서 폐색영역비용 값과 정합 보상값을 경험적으로 정하여왔다. 본 논문에서는 변이공간영상 방법에서 폐색영역비용과 정합보상값이 영상 잡음과 물체와 배경간의 차이에 영향을 받는 것을 이론적으로 분석하였다.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2005년도 추계학술대회 논문집
• /
• pp.191-196
• /
• 2005

In this study, to describe vehicle-track-bridge dynamic interaction phenomena with 1/4 vehicle model, nonlinear Hertzian contact spring and nonlinear contact damper are introduced. In this approach external loads acting on 1/4 vehicle model are self weight of vehicle and geometry information of running surface. The constraint equation on contact surface is implemented by Penalty method. Also, to improve the numerical stability and to maintain accuracy of solution, the artificial damper and the reaction from constraint violation are introduced. A nonlinear time integration method, in this study, Newmark method is adopted for both equations of vehicles and structure. And to reduce the error caused by inadequate time step size, adaptive time-stepping technique is partially introduced. As the nonlinear Hertzian contact spring has no resistance to tensile force, the bouncing phenomena of wheelset can be described. Thus, it is expected that more versatile dynamic interaction phenomena can be described by this approach and it can be applied to various railway dynamic problems.

• 반도체디스플레이기술학회지
• /
• 제12권3호
• /
• pp.13-18
• /
• 2013

Motion estimation is a key technique of modern video processing that significantly improves the coding efficiency significantly by exploiting the temporal redundancy between successive frames. Thread-level parallelism is a promising method to accelerate the motion estimation process for multithreading general-purpose processors. In this paper, we propose a parallel motion estimation algorithm which parallelizes the motion search process of the current H.264/AVC encoder. The proposed algorithm is implemented using the OpenMP application programming interface (API) and can be easily integrated into the current encoder. The experimental results show that the proposed parallel algorithm can reduce the processing time of the motion estimation up to 65.08% without any penalty in the rate-distortion (RD) performance.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2002년도 학술발표회 논문집(II)
• /
• pp.829-836
• /
• 2002

The paper presents a multi-objective genetic algorithm approach to the design of a water distribution network. The objectives considered are minimization of network cost and maximization of a reliability measure. In this study, a new reliability measure, called network resilience, is introduced. This measure mimics a designer's desire of providing excess power at nodes and designing reliable loops with practicable pipe diameters. The proposed method produces a set of Pareto-optimal solutions in the search space of cost and network resilience. Genetic algorithms are observed to be poor in handling constraints. To handle constraints in a better way, a constraint handling technique that does not require a penalty coefficient and applicable to water distribution systems is presented. The present model is applied to two example problems, which were widely reported. Pipe failure analysis carried out on some of the solutions obtained revealed that the network resilience based approach gave better results in terms of network reliability.

• 산업경영시스템학회지
• /
• 제22권49호
• /
• pp.115-123
• /
• 1999

An optimization problem is to select the best of many possible design alternatives in a complex design space. Genetic algorithms, one of the numerous techniques to search optimal solution, have been successfully applied to various problems (for example, parameter tuning in expert systems, structural systems with a mix of continuous, integer and discrete design variables) that could not have been readily solved with more conventional computational technique. But, conventional genetic algorithms are ill defined for two classes of problems, ie., penalty function and fitness scaling. Therefore, this paper develops Improved genetic algorithms(IGA) to solve these problems. As a case study, numerical examples are demonstrated to show the effectiveness of the Improved genetic algorithms.

• ETRI Journal
• /
• 제43권5호
• /
• pp.787-798
• /
• 2021

This paper applies learning automata to improve the performance of a Q-learning based random access channel (QL-RACH) scheme in a cellular machine-to-machine (M2M) communication system. A prioritized learning automata QL-RACH (PLA-QL-RACH) access scheme is proposed. The scheme employs a prioritized learning automata technique to improve the throughput performance by minimizing the level of interaction and collision of M2M devices with human-to-human devices sharing the RACH of a cellular system. In addition, this scheme eliminates the excessive punishment suffered by the M2M devices by controlling the administration of a penalty. Simulation results show that the proposed PLA-QL-RACH scheme improves the RACH throughput by approximately 82% and reduces access delay by 79% with faster learning convergence when compared with QL-RACH.

• ETRI Journal
• /
• 제41권5호
• /
• pp.619-625
• /
• 2019

In this paper, we propose an array pattern synthesis scheme using semidefinite programming (SDP) under array excitation power constraints. When an array pattern synthesis problem is formulated as an SDP problem, it is known that an additional rank-one constraint is generated inevitably and relaxed via semidefinite relaxation. If the solution to the relaxed SDP problem is not of rank one, then conventional SDP-based array pattern synthesis approaches fail to obtain optimal solutions because the additional rank-one constraint is not handled appropriately. To overcome this drawback, we adopted a bisection technique combined with a penalty function method. Numerical applications are presented to demonstrate the validity of the proposed scheme.