• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.11a
• /
• pp.471-474
• /
• 2005

원격 에이전트 서버는 에이전트 개발자가 제어권한올 가지지 못한다. 원격 에이전트 서버 중 일부가 다운된 상황에서는 에이전트를 정상적으로 동작시키기 어렵다. 이런 경우 일반적으로 이용하는 방법은 이동 에이전트가 각 에이전트 서버에 도착하면 복제본을 저장해 두는 것이다. 그러나 서버 고장 상태로부터 온라인 상태로 돌아올 때까지 복제본을 이용할 수 없게 되므로 적합한 해결책 되지 못한다. 본 논문은 에이전트 서버 고장시 이동 에이전트가 계속 동작할 수 있는 결함허용 처리기법올 제안한다. 제안 기법은 이동 에이전트를 생성하는 홈 에이전트 서버와 이동 에이전트가 방문한 이전 에이전트 서버에서 수행되며 원격 에이전트 서버에 고장이 발생한 경우 제안기법을 통해 결함허용을 처리하며 실험을 통해 제안 기법에서의 요소별 오버헤드를 분석한다.

• Journal of Information Processing Systems
• /
• v.1 no.1 s.1
• /
• pp.107-111
• /
• 2005

When a pedagogical agent system aims to provide students with interactive help, it needs to know what knowledge the student has and what goals the student is currently trying to achieve. That is, it must do both assessment and plan recognition. These modeling tasks involve a high level of uncertainty when students are allowed to follow various lines of reasoning and are not required to show all their reasoning explicitly. In this paper, the student model for interactive edutainment applications is proposed. This model is based on Bayesian Networks to expose constructs and parameters of rules and propositions pertaining to game and problem solving activities. This student model could be utilized as the emotion generation model for student and agent as well.

• Proceedings of the KIEE Conference
• /
• 2005.07b
• /
• pp.1371-1373
• /
• 2005

Nowadays, Intelligent Electronic Devices(IED) are widely used in power system. In order to improve the selectivity, sensitivity, and reliability of the power system composed by IEDs, A new fault tolerant IED system based on agent technique is presented in this paper. in the presented system, different with existing IED systems, the redundancies are drawn out from IED devices, and compose a standby backup system. And those redundancies are not specified to a particular type of IEDs. By using the agent technique, redundancies can download agents to modify their functions to fit different types of IEDs. As a conclusion, the reliability of the presented system is estimated in this paper.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.403-405
• /
• 2004

A service restoration is one of the most important missions in distribution system operation. This paper proposes a multi-agent system approach to distribution system restoration. Every relay is developed as an agent by adding its own intelligent, self-tuning and communication ability. Relay agent calculates and corrects its restoration index by itself through communication with neighboring agents and its own intelligence. The proposed algorithm is applied to a simple network to show how to calculate restoration index. Keywords Multi-Agent System, Service Restoration, Distribution Networks

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2019.05a
• /
• pp.3-4
• /
• 2019

본 논문에서는 복잡한 해상교통 환경 하에서도 해양 안전을 도모할 수 있는 강화학습 기반 지능형 선박 항해 에이전트 개발의 사전단계로서 기존의 강화학습 환경을 분석하였다. 강화학습 기반 접근법은 선박 항해 에이전트 스스로가 복잡하고 동적인 해상교통 환경을 이해하고 주어진 목표를 달성할 수 있도록 도와주는데, 이를 위해서는 에이전트 자신을 제외한 모든 사항들이 정의되는 환경을 보다 정확하고 효과적으로 개발하는 것이 매우 중요하다. 실제 해상교통 환경은 학습 환경으로의 모델링 및 에이전트 학습의 난이도가 매우 높은 환경으로 학습환경이 가질 수 있는 여러 속성들을 적절히 설정하여 선박 항해 에이전트의 활용 목적에 맞는 가성비 높은 환경을 구축하는 것이 바람직하다.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.1
• /
• pp.131-146
• /
• 2004

As Electronic Commerce(EC) has been emerged and has developed, many researchers have tried to establish EC framework for automated contract and negotiation using agent technologies. Traditional researches, however, often had limitations. They often enforced the user's participations during the automated contract process of agents. They also could only consider a few of the user's requirements for a specific goods and did not have supported the procedures and methodologies for making the best contract. In this paper, we propose business model on EC based on multiagents to overcome the defects of the previous researches. We apply CSP techniques to brokerage process to satisfy various preferential requirements from the user. We also propose efficient negotiation mechanism using negotiation model of game theory. The contract candidates automatically negotiate and mediate in terms of their benefits through the proposed negotiation mechanism. For the optimal brokerage and automated negotiation, the agents process activities for contract on three layers, which are called competition layer, constraint satisfaction layer and negotiation layer in the proposed model. We also design the message driven communication protocol to support the automated contract among the agents. Finally, we have implemented prototype systems applying the proposed model and have shown the various experimental results for efficiency of the proposed model.

• The Journal of Society for e-Business Studies
• /
• v.9 no.3
• /
• pp.155-167
• /
• 2004

At this time, the two main techniques for achieving web personalization assistant systems generally concern direct manipulation and software agents. While both direct manipulation and software agents are intended for permitting user to complete tasks rapidly, efficiently, and easily, their methodologies are different. The central debate involving these web personalization techniques originates from the amount of control that each allows to, or holds back from, the users. Direct manipulation can provide users with comprehensibel, predictable and controllable user interfaces that give them a feeling of accomplishnent and responsibility. On the other hand, the intelligent software components, the agents, can assist users with artificial intelligence by monitoring or retrieving personal histories or behaviors. In this empirical study, two web personalization assistant systems are evaluated. One of them, WebPersonalizer, is an agent based user personalization tool; the other, AntWorld, is a collaborative recommendation tool which provides direct manipulation interfaces. Through this empirical study, we have focused on two different paradigms as web personalization assistant systems : direct manipulation and software agents. Each approach has its own advantages and disadvantages. We also provide the experimental result that is worth referring for developers of electronic commerce system and suggest the methodologies for conveniently retrieving necessary information based on their personal needs.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.32-43
• /
• 2024

Game agents which are behavioral agent for game playing are a crucial component of game satisfaction. However it takes a lot of time and effort to create game agents for various game levels, environments, and players. In addition, when the game environment changes such as adding contents or updating characters, new game agents need to be developed and the development difficulty gradually increases. And it is important to have a game agent that can be customized for different levels of players. This is because a game agent that can play games of various levels is more useful and can increase the satisfaction of more players than a high-level game agent. In this paper, we propose a method for learning and controlling the level of play of game agents that can be rapidly developed and fine-tuned for various game environments and changes. At this time, reinforcement learning applies a policy-based distributed reinforcement learning method IMPALA for flexible processing and fast learning of various behavioral structures. Once reinforcement learning is complete, we choose actions by sampling based on Softmax-Temperature method. From this result, we show that the game agent's play level decreases as the Temperature value increases. This shows that it is possible to easily control the play level.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.3
• /
• pp.253-261
• /
• 2004

Game theory is mathematical analysis developed to study involved in making decisions. In 1928, Von Neumann proved that every two-person, zero-sum game with finitely many pure strategies for each player is deterministic. As well, 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. Not the rationality but through the process of Darwinian selection, a population of agents can evolve to an Evolutionary Stable Strategy (ESS) introduced by Maynard Smith. Keeping pace with these game theoretical studies, the first computer simulation of co-evolution was tried out by Hillis in 1991. Moreover, Kauffman proposed NK model to analyze co-evolutionary dynamics between different species. He showed how co-evolutionary phenomenon reaches static states and that these states are Nash equilibrium or ESS introduced in game theory. Since the studies about co-evolutionary phenomenon were started, however many other researchers have developed co-evolutionary algorithms, in this paper we propose Game theory based Co-Evolutionary Algorithm (GCEA) and confirm that this algorithm can be a solution of evolutionary problems by searching the ESS.To evaluate newly designed GCEA approach, we solve several test Multi-objective Optimization Problems (MOPs). From the results of these evaluations, we confirm that evolutionary game can be embodied by co-evolutionary algorithm and analyze optimization performance of GCEA by comparing experimental results using GCEA with the results using other evolutionary optimization algorithms.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.1
• /
• pp.87-94
• /
• 2007

Many researchers have modeled complicated problems with distributed AI technologies in distributed environment. Especially, intelligent agent technology we often proposed among researchers as efficient method to solve the complicated problems because agent technology makes possibile to share different problem solving abilities among processors in distributed processing environment. In this paper, we propose multiagent system model to solve distributed scheduling problem. At this point we apple CSP techniques to the multiagent model as individual problem solving ability of member agents. Scheduling problem is divided into subproblems according to constraints by distributed resources, then each agent solves its subproblem using CSP solver in the proposed model. This method improves scheduling efficiency. For meeting scheduling problem of case study, we show CSP modeling process and suggest problem solving procedure by multiagent system model.