• Journal of Institute of Control, Robotics and Systems
• /
• v.3 no.1
• /
• pp.52-60
• /
• 1997

In genetic algorithms(GA), a crossover is performed only at one or two places of a chromosome, and the fixed probabilities of crossover and mutation have been used during the entire generation. A GA with dynamic mutation is known to be superior to GAs with static mutation in performance, but so far no efficient dynamic mutation method has been presented. Accordingly in this paper, a GA is proposed to perform a uniform crossover based on the nucleotide(NU) concept, where DNA and RNA consist of NUs and also a concrete way to vary the probabilities of crossover and mutation dynamically for every generation is proposed. The efficacy of the proposed GA is demonstrated by its application to the unimodal, multimodal and nonlinear control problems, respectively. Simulation results show that in the convergence speed to the optimal value, the proposed GA was superior to existing ones, and the performance of GAs with varying probabilities of the crossover and the mutation improved as compared to GAs with fixed probabilities of the crossover and mutation. And it also shows that the NUs function as the building blocks and so the improvement of the proposed algorithm is supported by the building block hypothesis.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.5
• /
• pp.294-302
• /
• 2020

In a genetic algorithm, computer simulations are performed based on the natural evolution process of life, such as selection, crossover, and mutation. The genetic algorithm searches the approximate optimal solution by the parallel arrangement of Schema, which has a short definition length, low order, and high adaptability. This study examined the possibility of improving the efficiency of the optimal solution by considering the characteristics of the building block hypothesis, which are one of the key operating principles of a genetic algorithm. This study evaluated the efficiency of the optimization results according to the gene sequence for the implementation in solving problems. The optimal design problem of the water pipe was selected, and the genetic arrangement order reflected the engineering specificity by dividing into the existing, the network topology-based, and the flowrate-based arrangement. The optimization results with a flowrate-based arrangement were, on average, approximately 2-3% better than the other batches. This means that to increase the efficiency of the actual engineering optimization problem, a methodology that utilizes clear prior knowledge (such as hydraulic properties) to prevent such excellent solution characteristics from disappearing is essential. The proposed method will be considered as a tool to improve the efficiency of large-scale water supply network optimization in the future.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• /
• pp.679-682
• /
• 2003

In this paper, we propose a Genetic Algorithm (GA) using symbiotic evolutionary viruses. Our GA is based on both the building block hypothesis and the virus theory of evolution. The proposed GA aims to control a destruction of building blocks by discovering, keeping, and propagating of building blocks based on virus operation. Concretely, we prepare the group of individuals and the group of viruses. In our GA, the group of individuals searches solutions and the group of viruses searches building blocks. These searches done based on the symbiotic relation of both groups. Also, our GA has two types of virus evaluation techniques. One is that each virus is evaluated by the difference of the fitness of an individual between before and after infection of virus. Another is that all viruses aye evaluated by the difference of the fitness of an individual between before and after infection of all viruses. Furthermore, we applied the proposed GA to the minimum value search problem of a test function which has some local solutions far from the optimal solution. And, we discuss a difference of behaviors of the proposed GA based on each virus evaluation techniques.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• /
• pp.77-80
• /
• 1998

The theoretical foundations of simple genetic algorithm(SGA) are the Schema Theorem and the Building Block Hypothesis. Although SGA does well in many applications as an optimization method, still it does not guarantee the convergence of a global optimum in GA-hard problems and deceptive problems. Therefore as an alternative scheme, there is a growing interest in a co-evolutionary system, where two populations constantly interact and cooperate each other. In this paper we show why the co-evolutionary algorithm works better than SGA in terms of an extended schema theorem. Also the experimental results show a co-evolutionary algorithm works well in optimization problems.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• /
• pp.553-559
• /
• 1998

Simple Genetic Algorithm(SGA) proposed by J. H. Holland is a population-based optimization method based on the principle of the Darwinian natural selection. The theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. Although GA goes well in many applications as an optimization method, still it does not guarantee the convergence to a global optimum in some problems. In designing intelligent systems, specially, since there is no deterministic solution, a heuristic trial-and error procedure is usually used to determine the systems' parameters. As an alternative scheme, therefore, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve. In this paper we review the existing co-evolutionary algorithms and propose co-evolutionary schemes designing intelligent systems according to the relation between the system's components.

• Proceedings of the IEEK Conference
• /
• /
• pp.1013-1016
• /
• 1999

Simple Genetic Algorithm(SGA) proposed by J. H. Holland is a population-based optimization method based on the principle of the Darwinian natural selection. The theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. Although GA does well in many applications as an optimization method, still it does not guarantee the convergence to a global optimum in GA-hard problems and deceptive problems. Therefore as an alternative scheme, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve. In this paper we propose an extended schema theorem associated with a schema co-evolutionary algorithm(SCEA), which explains why the co-evolutionary algorithm works better than SGA. The experimental results show that the SCEA works well in optimization problems including deceptive functions.

• 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 Korean Society for Industrial and Applied Mathematics
• /
• v.3 no.1
• /
• pp.115-125
• /
• 1999

Simple Genetic Algorithm(SGA) proposed by J. H. Holland is a population-based optimization method based on the principle of the Darwinian natural selection. The theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. Although GA does well in many applications as an optimization method, still it does not guarantee the convergence to a global optimum in some problems. In designing intelligent systems, specially, since there is no deterministic solution, a heuristic trial-and error procedure is usually used to determine the systems' parameters. As an alternative scheme, therefore, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve. In this paper we review the existing co-evolutionary algorithms and propose co-evolutionary schemes designing intelligent systems according to the relation between the system's components.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• /
• pp.123-126
• /
• 1998

The theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. In the Meaning of these foundational concepts, simple genetic algorithm(SGA) allocate more trials to the schemata whose average fitness remains above average. Although SGA does well in many applications as an optimization method, still it does not guarantee the convergence of a global optimum. Therefore as an alternative scheme, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve in contrast with traditional single population evolutionary algorithms. In this paper, we propose a new design method of an optimal fuzzy logic controller using co-evolutionary concept. In general, it is very difficult to find optimal fuzzy rules by experience when the input and/or output variables are going to increase. So we propose a co-evolutionary method finding optimal fuzzy rules. Our algorithm is that after constructing two population groups m de up of rule vase and its schema, by co-evolving these two populations, we find optimal fuzzy logic controller. By applying the proposed method to a path planning problem of autonomous mobile robots when moving objects exist, we show the validity of the proposed method.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.2 no.1
• /
• pp.95-110
• /
• 1998

Evolutionary Algorithms (EAs) are population-based optimization methods based on the principle of Darwinian natural selection. The representative methodology in EAs is genetic algorithm (GA) proposed by J. H. Holland, and the theoretical foundations of GA are the Schema Theorem and the Building Block Hypothesis. In the meaning of these foundational concepts, simple genetic algorithm (SGA) allocate more trials to the schemata whose average fitness remains above average. Although SGA does well in many applications as an optimization method, still it does not guarantee the convergence of a global optimum in GA-hard problems and deceptive problems. Therefore as an alternative scheme, there is a growing interest in a co-evolutionary system, where two populations constantly interact and co-evolve in contrast with traditional single population evolutionary algorithm. In this paper we show why the co-evolutionary algorithm works better than SGA in terms of an extended schema theorem. And predator-prey co-evolution and symbiotic co-evolution, typical approaching methods to co-evolution, are reviewed, and dynamic fitness landscape associated with co-evolution is explained. And the experimental results show a co-evolutionary algorithm works well in optimization problems even though in deceptive functions.