• 대한산업공학회지
• /
• 제28권1호
• /
• pp.87-98
• /
• 2002

A competitive coevolutionary algorithm is a probabilistic search method that imitates coevolution process through evolutionary arms race. The algorithm has been used to solve adversarial problems. In the algorithms, the selection of competitors is needed to evaluate the fitness of an individual. The goal of this study is to compare and analyze several competition strategies in terms of solution quality, convergence speed, balance between competitive coevolving species, population diversity, etc. With two types of test-bed problems, game problems and solution-test problems, extensive experiments are carried out. In the game problems, sampling strategies based on fitness have a risk of providing bad solutions due to evolutionary unbalance between species. On the other hand, in the solution-test problems, evolutionary unbalance does not appear in any strategies and the strategies using information about competition results are efficient in solution quality. The experimental results indicate that the tournament competition can progress an evolutionary arms race and then is successful from the viewpoint of evolutionary computation.

• Nuclear Engineering and Technology
• /
• 제56권9호
• /
• pp.3463-3471
• /
• 2024

The physical protection systems of Nuclear Power Plant are utilized to safeguard targets against intrude by attacker. As the methods employed by attackers to intrude Nuclear Power Plant become increasingly complex and diverse, there is an urgent need to identify optimal defense strategies to interrupt adversary intrusions. This paper focuses on studying the defense of security personnel against adversary intrusions and utilizes an evolutionary game approach to select the optimal defense decisions for physical protection systems. Under the assumption of bounded rationality for both the attacker and defender, the paper constructs replication dynamic equations for attack and defense strategies, investigating the process of strategy selection and the stability of evolution. Finally, a minimal model is proposed to validate the feasibility of utilizing the evolutionary game model for defense strategy selection.

• 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.

• Industrial Engineering and Management Systems
• /
• 제14권1호
• /
• pp.44-57
• /
• 2015

In recent years, for the purpose of solving the problem regarding environment protection and resource saving, certain measures and policies have been promoted to establish a reverse supply chains (RSCs) with material flows from collection of used products to reuse the recycled parts in production of products. It is necessary to analyze behaviors of RSC members to determine the optimal operation. This paper discusses a RSC with a retailer and a manufacturer and verifies the behavior strategies of RSC members which may change over time in response to changes parameters related to the recycling promotion activity in RSC. A retailer takes two behaviors: cooperation/non-cooperation in recycling promotion activity. A manufacturer takes two behaviors: monitoring/non-monitoring of behaviors of the retailer. Evolutionary game theory combining the evolutionary theory of Darwin with game theory is adopted to clarify analytically evolutionary outcomes driven by a change in each behavior of RSC members over time. The evolutionary stable strategies (ESSs) for RSC members' behaviors are derived by using the replicator dynamics. The analysis numerically demonstrates how parameters of the recycling promotion activity: (i) sale promotion cost, (ii) monitoring cost, (iii) compensation and (iv) penalty cost affect the judgment of ESSs of behaviors of RSC members.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.549-550
• /
• 2021

본 논문에서는 다양한 돌연변이 전략을 인식하기 위해 MUDE (Uniform Local Search)를 사용한 다중 인구 차등 진화 알고리즘을 제안한다. MUDE에서 집단은 진화 비율(DE/rand/1 및 DE/current-to-rand/1)에 따라 서로 다른 돌연변이 전략을 수행하는 다른 집단 크기를 가진 여러 하위 집단으로 나뉜다. 인구의 다양성을 개선하기 위해 정보는 소프트 아일랜드 모델에 의해 하위 인구 간에 마이그레이션된다.

• 한국지능시스템학회논문지
• /
• 제14권3호
• /
• pp.253-261
• /
• 2004

게임 이론은 의사 결정 문제와 관련 된 연구와 함께 정립 된 수학적 분석법으로써 1928년 Von Neumann이 유한개의 순수전략이 존재하는 2인 영합게임은 결정적(deterministic)이라는 것을 증명함으로써 수학적 기반을 정립하였고 50년대 초, Nash는 Von Neumann의 이론을 일반화하는 개념을 제안함으로써 현대적 게임이론의 장을 열었다. 이후 진화 생물학 연구자들에 의해 고전적인 게임 이론의 가정에 해당하는 참가자들의 합리성(rationality) 대신 다윈 선택(Darwinian selection)에 의해 게임의 해를 탐색하는 것이 가능하다는 것이 밝혀지게 되었고 진화 생물학자 Maynard Smith에 의해 진화적 안정 전략(Evolutionary Stable Strategy: ESS)의 개념이 정립되면서 현대적 게임 이론으로써 진화적 게임 이론이 체계화 되었다. 한편 이와 같은 진화적 게임 이론에 관한 연구와 함께 생태계의 공진화를 이용한 컴퓨터 시뮬레이션이 1991년 Hillis에 의해 처음으로 시도되었으며 Kauffman은 다른 종들 간의 공진화적 동역학(dynamics)을 분석하기 위한 NK 모델을 제안하였다. Kauffman은 이 모델을 이용하여 공진화 현상이 어떻게 정적 상태(static state)에 이르며 이 상태들은 게임 이론에서 소개되어진 내쉬 균형이나 ESS에 해당한다는 것을 보여주었다. 이후, 몇몇 연구자들 게임 이론과 진화 알고리즘에 기반한 연산 모델들을 제시해 왔으나 실용적인 문제의 적용에 대한 연구는 아직 미흡한 편이다. 이에 본 논문에서는 게임 이론에 기반 한 공진화 알고리즘을(Game theory based Co-Evolutionary Algorithm: GCEA) 제안하고 이 알고리즘을 이용하여 공진화적인 문제들을 효과적으로 해결할 수 있음을 확인하는 것을 목표로 한다.

• 한국경영과학회지
• /
• 제25권4호
• /
• pp.67-80
• /
• 2000

Symbiotic evolutionary algorithms, also called cooperative coevolutionary algorithms, are stochastic search algorithms that imitate the biological coevolution process through symbiotic interactions. In the algorithms, the fitness evaluation of an individual required first selecting symbiotic partners of the individual. Several partner selection strategies are provided. The goal of this study is to analyze how much partnering strategies can influence the performance of the algorithms. With two types of test-bed problems: the NKC model and the binary string covering problem, extensive experiments are carried out to compare the performance of partnering strategies, using the analysis of variance. The experimental results indicate that there does not exist statistically significant difference in their performance.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 추계학술대회 학술발표 논문집
• /
• pp.420-426
• /
• 1998

Evolutionary algorithms are probabilistic optimization algorithms based on the model of natural evolution. Recently the efforts to improve the performance of evolutionary algorithms have been made extensively. In this paper, we introduce the research for improving the convergence rate and search faculty of evolution algorithms by using reinforcement learning. After providing an introduction to evolution algorithms and reinforcement learning, we present adaptive genetic algorithms, reinforcement genetic programming, and reinforcement evolution strategies which are combined with reinforcement learning. Adaptive genetic algorithms generate mutation probabilities of each locus by interacting with the environment according to reinforcement learning. Reinforcement genetic programming executes crossover and mutation operations based on reinforcement and inhibition mechanism of reinforcement learning. Reinforcement evolution strategies use the variances of fitness occurred by mutation to make the reinforcement signals which estimate and control the step length.

• 한국산업융합학회 논문집
• /
• 제27권4_2호
• /
• pp.909-923
• /
• 2024

This study aims to classify the types of financial super apps and analyzes their evolution and growth paths by type. Super apps, which provide various services on a single platform, are gaining attention as a key strategy for digital transformation in the financial sector. By adopting the grounded theory methodology, this research has categorized financial super apps into three types: "lifestyle financial super app", "integrated financial super app", and "universal financial super app". Ansoff Matrix was used as a theoretical framework to understand how each type of super app grew and evolved through various strategies. Our analysis revealed that super apps of each type grew using a different mix of 'market penetration', 'product development', 'mark et development', and 'diversification' strategies, with each mix showcasing a distinct evolutionary path. The findings of this study are expected to enhance understanding of financial super app typology and evolutionary trajectories, contributing to the development of practical strategies, such as channel optimization for financial super apps in the future.

• International Journal of Control, Automation, and Systems
• /
• 제2권4호
• /
• pp.463-474
• /
• 2004

Game theory is a method of mathematical analysis developed to study the decision making process. In 1928, Von Neumann mathematically proved that every two-person, zero-sum game with many pure finite strategies for each player is deterministic. In the early 50's, Nash presented another concept as the basis for a generalization of Von Neumann's theorem. Another central achievement of game theory is the introduction of evolutionary game theory, by which agents can play optimal strategies in the absence of rationality. Through the process of Darwinian selection, a population of agents can evolve to an Evolutionary Stable Strategy (ESS) as introduced by Maynard Smith in 1982. Keeping pace with these game theoretical studies, the first computer simulation of coevolution was tried out by Hillis. Moreover, Kauffman proposed the NK model to analyze coevolutionary dynamics between different species. He showed how coevolutionary phenomenon reaches static states and that these states are either Nash equilibrium or ESS in game theory. Since studies concerning coevolutionary phenomenon were initiated, there have been numerous other researchers who have developed coevolutionary algorithms. In this paper we propose a new coevolutionary algorithm named Game theory based Coevolutionary Algorithm (GCEA) and we confirm that this algorithm can be a solution of evolutionary problems by searching the ESS. To evaluate this newly designed approach, we solve several test Multiobjective Optimization Problems (MOPs). From the results of these evaluations, we confirm that evolutionary game can be embodied by the coevolutionary algorithm and analyze the optimization performance of our algorithm by comparing the performance of our algorithm with that of other evolutionary optimization algorithms.