• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.531-531
• /
• 2000

The challenge is to control unstable nonlinear dynamic systems using only sparse feedback from the environment concerning its performance. The design of such controllers can be achieved by evolving neural networks. An evolutionary approach to train neural networks in realtime is proposed. Evolutionary strategies adapt the weights of neural networks and the threshold values of neuron's synapses. The proposed method has been successfully implemented for pole balancing problem.

• 한국정보과학회논문지:정보통신
• /
• 제32권3호
• /
• pp.301-309
• /
• 2005

시스템 호출 감사자료기반 기계학습기법을 사용한 프로그램 행위 학습방법은 효과적인 호스트 기반 침입탐지 방법이며, 규칙 학습, 신경망, 통계적 방법, 은닉 마크로프 모델 등의 방법이 대표적이다. 그 중에서 신경망은 시스템 호출 시퀀스를 학습하는데 있어 적합하다고 알려져 있는데, 실제 문제에 적용하여 좋은 성능을 내기 위해서는 그 구조를 결정하는 것이 중요하다 하지만 보통의 신경망은 그 구조를 찾기 위한 방법이 알려져 있지 않아 침입탐지에 효과적인 구조를 찾기 위해서는 많은 시간이 요구된다. 본 논문에서는 기존 신경망 기반 침입탐지시스템의 단점을 보완하고 성능을 향상시키기 위해 진화신경망을 이용한 방법을 제안한다. 진화 신경망은 신경망의 구조와 가중치를 동시에 학습하기 때문에 일반 신경망보다 빠른 시간에 더 좋은 성능의 신경망을 얻을 수 있다는 장점이 있다. 1999년의 DARPA IDEVAL 자료로 실험한 결과 기존의 연구보다 좋은 탐지율을 보여 진화신경망이 침입탐지에 효과적임을 확인할 수 있었다.

• International Journal of Aeronautical and Space Sciences
• /
• 제3권1호
• /
• pp.50-60
• /
• 2002

A new co-evolutionary algorithm, of which the convergence speed is accelerated by neural networks, is proposed and verified in this paper. To reduce computational load required for co-evolutionary optimization processes, the cost function and constraint information is stored in the neural networks, and the extra offspring group, whose cost is computed by the neural networks, is generated. It increases the offspring population size without overloading computational effort; therefore, the convergence speed is accelerated. The proposed algorithm is applied to attitude control design of flexible satellites, and it is verified by computer simulations and experiments using a torque-free air bearing system.

• 한국지능시스템학회논문지
• /
• 제11권5호
• /
• pp.400-405
• /
• 2001

This paper addresses an effective approach of training neural networks classifiers for credit card fraud detection. The proposed approach uses evolutionary programming to trails the neural networks classifiers based on maximization of the detection rate of fraudulent usages on some ranges of the rejection rate, loot minimization of mean square error(MSE) that Is a common criterion for neural networks learning. This approach enables us to get classifier of satisfactory performance and to offer a directive method of handling various conditions and performance measures that are required for real fraud detection applications in the classifier training step. The experimental results on "real"credit card transaction data indicate that the proposed classifiers produces classifiers of high quality in terms of a relative profit as well as detection rate and efficiency.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.43.2-43
• /
• 2001

In this paper, we consider the problems of the evolutionary designed neural networks structures by genetic algorithm. Neural networks has been applied to various application fields since back-propagation algorithm was proposed, e.g. function approximation, pattern or character recognition and so on. However, one of difficulties to use the neural networks. It is how to design the structure of the neural network. Researchers and users design networks structures and training parameters such as learning rate and momentum rate and so on, by trial and error based on their experiences. In the case of designing large scales neural networks, it is very hard work for manually design by try and error. For this difficulty, various structural learning algorithms have been proposed. Especially, the technique of using genetic algorithm for networks structures design has been ...

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1997년도 추계학술대회 학술발표 논문집
• /
• pp.111-114
• /
• 1997

Usually, Evolutionary Algorithms are considered more efficient for optimal system design, However, the performance of the system is determined by fitness function and system environment. In this paper, in order to overcome the limitation of the performance by this factor, we propose a co-evolutionary method that two populations constantly interact and coevolve. In this paper, we apply coevolution to neural network's evolving. So, one population is composed of the structure of neural networks and other population is composed of training patterns. The structure of neural networks evolve to optimal structure and, at the same time, training patterns coevolve to feature patterns. This method prevent the system from the limitation of the performance by random design of neural network structure and inadequate selection of training patterns. In this time neural networks are trained by evolution strategies that are able to apply to the unsupervised learning. And in the coding of neural networks, we propose the method to maintain nonredundancy and character preservingness that are essential factor of genetic coding. We show the validity and the effectiveness of the proposed scheme by applying it to the visual servoing of RV-M2 robot manipulators.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제52권7호
• /
• pp.424-433
• /
• 2003

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제52권7호
• /
• pp.424-424
• /
• 2003

Evolutionary design related to the optimal design of Polynomial Neural Networks (PNNs) structure for model identification of complex and nonlinear system is studied in this paper. The PNN structure is consisted of layers and nodes like conventional neural networks but is not fixed and can be changable according to the system environments. three types of polynomials such as linear, quadratic, and modified quadratic is used in each node that is connected with various kinds of multi-variable inputs. Inputs and order of polynomials in each node are very important element for the performance of model. In most cases these factors are decided by the background information and trial and error of designer. For the high reliability and good performance of the PNN, the factors must be decided according to a logical and systematic way. In the paper evolutionary algorithm is applied to choose the optimal input variables and order. Evolutionary (genetic) algorithm is a random search optimization technique. The evolved PNN with optimally chosen input variables and order is not fixed in advance but becomes fully optimized automatically during the identification process. Gas furnace and pH neutralization processes are used in conventional PNN version are modeled. It shows that the designed PNN architecture with evolutionary structure optimization can produce the model with higher accuracy than previous PNN and other works.

• 한국멀티미디어학회논문지
• /
• 제12권2호
• /
• pp.290-299
• /
• 2009

진화하는 인공신경망은 인공지능분야와 게임 NPC의 지능 설계 분야에서 새롭게 각광을 받고 있다. 하지만 진화하는 인공신경 망을 이용하여 게임 NPC의 지능을 설계할 때 인공신경 망의 구조가 복잡함에 따라 진화와 평가에 필요한 연산량이 크며 또한 적절한 적합도 함수를 설계하지 못하면 지능적인 NPC를 설계할 수 없는 등의 문제점을 가지고 있다. 본 논문에서는 이러한 문제들을 해결하고자 동적 상태 진화 인공신경망을 제안한다. 동적 상태 진화 인공신경망은 전통적인 진화하는 인공신경망 알고리즘에 기반하여 진화 과정에서 신경망의 신경세포들 사이의 시냅스를 제거(disabled) 하거나 고정(fixed)시키는 방법을 통하여 진화와 평가과정에 소모되는 연산량을 줄이는 알고리즘이다. 본 논문은 Darwin Platform 을 테스트 베드로 축구게임 NPC의 지능 설계를 통하여 제안하는 방법의 유용성을 검증한다.

• 제어로봇시스템학회논문지
• /
• 제5권3호
• /
• pp.281-287
• /
• 1999

In this paper, a design method based on evolutionary programming for feedforward neural networks which have a single hidden layer is presented. By using an evolutionary programming, the network parameters such as the network structure, weight, slope of sigmoid functions and bias of nodes can be acquired simultaneously. To check the effectiveness of the suggested method, two numerical examples are examined. The performance of the identified network is demonstrated.