• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.11
• /
• pp.75-83
• /
• 1999

We propose a PfSGA(parameter-free species genetic algorithm) to learn the topological structure of MLP classifiers being adequate to given applications. The PfSGA is a combinational method of SGA(species genetic algorithm) and PfGA(parameter-free genetic algorithm). In SGA, we divide the total search space into several subspaces(species) according to the number of hidden units, and reduce the unnecessary search by eliminating the low promising species from the evolutionary process. However the performances of SGA classifiers are readily affected by the values of parameters such as mutation ratio and crossover ratio. In this paper, therefore, we combine SGA with PfGA, for which it is not necessary to determine the learning parameters. Experimental results on benchmark data and sign language words show that PfSGA can reduce the learning time of SGA and is not affected by the selection parameter values on structural learning. The results also show that PfSGA is more efficient than the exisiting methods in the aspect of misclassification ratio, learning rate, and complexity of MLP structure.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.12 no.3
• /
• pp.82-88
• /
• 2003

SGA(Single Genetic Algorithm) is a heuristic global optimization method based on the natural characteristics and uses many populations and stochastic rules. Therefore SGA needs many function evaluations and takes much time for convergence. In order to solve the demerits of SGA, ${\mu}GA$(Micro-Genetic Algorithm) has recently been developed. In this study, ${\mu}GA$ which have small populations and fast convergence rate, was applied to structural optimization with discrete or integer variables such as 3, 10 and 25 bar trusses. The optimized results of ${\mu}GA$ were compared with those of SGA. Solutions of ${\mu}GA$ for structural optimization were very similar or superior to those of SGA, and faster convergence rate was obtained. From the results of examples, it is found that ${\mu}GA$ is a suitable and very efficient optimization algorithm for structural design.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2003.04a
• /
• pp.9-14
• /
• 2003

SGA (Single Genetic Algorithm) is a heuristic global optimization method based on the natural characteristics and uses many populations and stochastic rules. Therefore SGA needs many function evaluations and takes much time for convergence. In order to solve the demerits of SGA, $\mu$GA(Micro-Genetic Algorithm) has recently been developed. In this study, $\mu$GA which have small populations and fast convergence rate, was applied to structural optimization with discrete or integer variables such as 3, 10 and 25 bar trusses. The optimized results of $\mu$GA were compared with those of SGA. Solutions of $\mu$GA for structural optimization were very similar or superior to those of SGA, and faster convergence rate was obtained. From the results of examples, it is found that $\mu$GA is a suitable and very efficient optimization algorithm for structural design.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.63-66
• /
• 2004

The SGA(stochastic gradient ascent) algorithm is one of the most important tools in the area of reinforcement learning, and has been applied to a wide range of practical problems. In particular, this learning method was successfully applied by Kimura et a1. [1] to the control of a simple creeping robot which has finite number of control input choices. In this paper, we considered the application of the SGA algorithm to Kimura's robot control problem for the case that the control input is not confined to a finite set but can be chosen from a infinite subset of the real numbers. We also developed a MATLAB-based robot animation program, which showed the effectiveness of the training algorithms vividly.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.5
• /
• pp.616-623
• /
• 1998

Holland가 제안한 단순 유전자 알고리즘은 다원의 자연선택설을 기본으로 한 군 기반의 최적화 방법으로서, 이론적 기반으로는 스키마 정리와 빌딩블록 가설이 있다. 단순 유전자 알고리즘(SGA)이 이러한 이론적 기반에도 불구하고 여전히 일부 문제에 있어서 최적해로의 수렴을 보장하지 못하고 있다. 따라서 최근에 두 개의 집단이 서로 상호작용을 하며 진화하는 공진화 방법에 의해 이러한 문제를 해결하려고 하는데 많은 관심이 모아지고 있다. 본 논문에서는 이러한 공진화 방법이 잘 동작하는지에 대한 이론적 기반으로 확장 스키마 정리를 제안하고, SGA에서는 해결하지 못하는 최적화 문제, 예를 들면 deceptive function,에서 SGA와 공진화에 의한 방법을 비교함으로써 확장된 스키마 정리의 유효성을 확인한다.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.19 no.1 s.71
• /
• pp.27-37
• /
• 2006

A Metropolis genetic algorithm (MGA) is developed and applied for the structural design optimization. In MGA, favorable features of Metropolis criterion of simulated annealing (SA) are incorporated in the reproduction operations of simple genetic algorithm (SGA). This way, the MGA maintains the wide varieties of individuals and preserves the potential genetic information of early generations. Consequently, the proposed MGA alleviates the disadvantages of premature convergence to a local optimum in SGA and time consuming computation for the precise global optimum in SA. Performances and applicability of MGA are compared with those of conventional algorithms such as Holland's SGA, Krishnakumar's micro GA, and Kirkpatrick's SA. Typical numerical examples are used to evaluate the computational performances, the favorable features and applicability of MGA. The effects of population sizes and maximum generations are also evaluated for the performance reliability and robustness of MGA. From the theoretical evaluation and numerical experience, it is concluded that the proposed MGA Is a reliable and efficient tool for structural design optimization.

• Composites Research
• /
• v.21 no.1
• /
• pp.30-39
• /
• 2008

In this paper, a parallel micro genetic algorithm was utilized in the optimal design of composite structures instead of a conventional genetic algorithm(SGA). Micro genetic algorithm searches the optimal design variables with only 5 individuals. The diversities from the nominal convergence and the re-initialization processes make micro genetic algorithm to find out the optimums with such a small population size. Two different composite structure optimization problems were proposed to confirm the efficiency of micro genetic algorithm compared with SGA. The results showed that micro genetic algorithm can get the solutions of the same level of SGA while reducing the calculation costs up to 70% of SGA. The composite laminated structure optimization under the load uncertainty was conducted using micro genetic algorithm. The result revealed that the design variables regarding the load uncertainty are less sensitive to load variation than that of fixed applied load. From the above-mentioned results, we confirmed micro genetic algorithm as a optimization method of composite structures is efficient.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.2
• /
• pp.287-295
• /
• 2010

In this paper, we introduce a novel approach to optimization algorithm which is a distributed Mean field Genetic algorithm (MGA) implemented in MPI(Message Passing Interface) environments. Distributed MGA is a hybrid algorithm of Mean Field Annealing(MFA) and Simulated annealing-like Genetic Algorithm(SGA). The proposed distributed MGA combines the benefit of rapid convergence property of MFA and the effective genetic operations of SGA. The proposed distributed MGA is applied to the channel routing problem, which is an important issue in the automatic layout design of VLSI circuits. Our experimental results show that the composition of heuristic methods improves the performance over GA alone in terms of mean execution time. It is also proved that the proposed distributed algorithm maintains the convergence properties of sequential algorithm while it achieves almost linear speedup as the problem size increases.

• The KIPS Transactions:PartB
• /
• v.14B no.5
• /
• pp.369-376
• /
• 2007

This paper proposes a hybrid genetic algorithm(HGA) for the classifier ensemble selection. HGA is added a local search operation for increasing the fine-turning of local area. This paper apply hybrid and simple genetic algorithms(SGA) to the classifier ensemble selection problem in order to show the superiority of HGA. And this paper propose two methods(SSO: Sequential Search Operations, CSO: Combinational Search Operations) of local search operation of hybrid genetic algorithm. Experimental results show that the HGA has better searching capability than SGA. The experiments show that the CSO considering the correlation among classifiers is better than the SSO.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2002.04a
• /
• pp.225-232
• /
• 2002

Simple genetic algorithm(SGA) has been used to optimize a lot of structural optimization problems because it can optimize non-linear problems and obtain the global solution. But, because of large evolving populations during many generations, it takes a long time to calculate fitness. Therefore this paper applied micro-genetic algorithm(μ -GA) to structural optimization and compared results of μ -GA with results of SGA. Additionally, the Paper applied μ -GA to gate optimization problem for injection molds by using simulation program CAPA.