• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.4
• /
• pp.361-370
• /
• 2023

Swarm has recently become a critical component of offensive and defensive systems. Multi-agent reinforcement learning(MARL) empowers swarm systems to handle a wide range of scenarios. However, the main challenge lies in MARL's scalability issue - as the number of agents increases, the performance of the learning decreases. In this study, transfer learning is applied to advanced MARL algorithm to resolve the scalability issue. Validation results show that the training efficiency has significantly improved, reducing computational time by 31 %.

• International Journal of Contents
• /
• v.16 no.1
• /
• pp.25-33
• /
• 2020

We develop a simulation method that affords very high variability of virtual pedagogical situations involving many independent plans, still achieves authoring (or implementation) scalability. While each individual plan would be coherently drawn up by an agent for its respective goal, those independently-made plans might be coincidentally intertwined in their execution. The inevitable non-determinism involved in this multi-event plan encompassing pre-planned and unforeseen events is resolved by (multi-phase) dynamic planning and articulated sequencing of events in contrast to static planning and monolithic authoring in conventional narrative systems. Connections between events are dictated by their associated rules and their actual connections are dynamically determined in execution time by current conditions of background-world. This unified connection scheme across pre-planned and unforeseen events allows a multi-plan, multi-agent situation to be coherently planned and executed in a global scale. To further the variability of a situation, the inter-event coupling is made in a fine level of action along with a limited episteme of each agent involved. We confirm analytically the viability of our approach with respect to the situation variability and authoring scalability, and demonstrate its practicality with an implementation of a composite situation.

• The Journal of Information Systems
• /
• v.14 no.3
• /
• pp.89-96
• /
• 2005

Though studies on negotiations have actively been conducted in the field of e-Commerce so far, some problems have yet to be solved and many application fields of negotiation systems exist. Currently, many businesses shift from various fields to e-Commerce market, due to recent social changes and expansion of e-Commerce market; however, we need to develop the study of automated agents to resolve the issue of negotiation, which is an essential element in the e-Commerce, by minimizing human interference under the e-Commerce environment. In this study, we intend to propose a new negotiation protocol considering the negotiation alternatives through continuous negotiation rounds in relation to an automated negotiation issue whose participants are one to N (seller to buyers). We also present an agent-facilitated negotiation methodology by which negotiation alternative generation process is automated, in consideration of buyer's negotiation attributes and strategies in the negotiation system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.7C
• /
• pp.997-1006
• /
• 2004

The rapid expansion of the Internet has provided users with a diverse range of services. Most Internet users create many different IDs and passwords to subscribe to various Internet services. Thus, the SSO system has been proposed to supplement vulnerable security that may arise from inefficient management system where administrators and users manage a number of ms. The SSO system can provide heightened efficiency and security to users and administrators. Recently commercialized SSO systems integrate a single agent with the broker authentication model. However, this hybrid authentication system cannot resolve problems such as those involving user pre-registration and anonymous users. It likewise cannot provide non-repudiation service between joining objects. Consequently, the hybrid system causes considerable security vulnerability. Since it cannot provide security service for the agent itself, the user's private information and SSO system may have significant security vulnerability. This paper proposed an authentication model that integrates a broker authentication model, out of various authentication models of the SSO system, with a multi-agent system. The proposed method adopts a secure multi-agent system that supplements the security vulnerability of an agent applied to the existing hybrid authentication system. The method proposes an SSO authentication model that satisfies various security requirements not provided by existing broker authentication models and hybrid authentication systems.

• The Journal of Information Systems
• /
• v.14 no.2
• /
• pp.55-74
• /
• 2005

In order to purchase a product cheaper, a lot of customers have been trying to search one or more marketplaces. Ever since the commercial use of the Internet, several types of marketplaces have been operating successfully on the Internet. Some of them are online shopping malls, auction markets, and group-buying markets. They have the price settlement mechanisms of their own. Online shopping malls where many stores are located support a customer to purchase the product that matches his/her requests such as price, function, design, and so forth. In online auction market, a customer can buy the product by making bids sequentially and competitively until a final price is reached. In online group-buying market, a customer can purchase the product by aggregating the orders from several buyers so that cheaper prices can be negotiated. The cheaper customers could purchase the same product item, the more satisfied they would be. However, it is very difficult for the customer to determine the marketplace to purchase, considering different kinds of marketplaces at the same time. Even though the purchasing price is cheapest in one marketplace, it is very difficult for customers to convince it the cheapest for all marketplaces. Therefore, rules and methods have been developed for purchase decision making in multiple marketplaces to reach the optimal purchase decision as a whole. They can maximize customer's utility and resolve the conflicts with other marketplaces through multi-agent negotiation.

• Journal of Intelligence and Information Systems
• /
• v.13 no.3
• /
• pp.119-140
• /
• 2007

Recommender systems enable a user to decide which information is interesting and valuable in our world of information overload. As the recent studies of distributed computing environment have been progressing actively, recommender systems, most of which were centralized, have changed toward a peer-to-peer approach. Collaborative Filtering (CF), one of the most successful technologies in recommender systems, presents several limitations, namely sparsity, scalability, cold start, and the shilling problem, in spite of its popularity. The move from centralized systems to distributed approaches can partially improve the issues; distrust of recommendation and abuses of personal information. However, distributed systems can be vulnerable to attackers, who may inject biased profiles to force systems to adapt their objectives. In this paper, we consider both effective CF in P2P environment in order to improve overall performance of system and efficient solution of the problems related to abuses of personal data and attacks of malicious users. To deal with these issues, we propose a multi-agent framework for a distributed CF focusing on the trust relationships between individuals, i.e. web of trust. We employ an agent-based approach to improve the efficiency of distributed computing and propagate trust information among users with effect. The experimental evaluation shows that the proposed method brings significant improvement in terms of the distributed computing of similarity model building and the robustness of system against malicious attacks. Finally, we are planning to study trust propagation mechanisms by taking trust decay problem into consideration.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.2
• /
• pp.243-253
• /
• 2019

First steps of Planetary exploration are usually conducted with the use of autonomous rovers. These rovers are capable of finding its own path and perform experiments about the planet's surface. This paper makes a proposal for a multi-agent system which effectively take the advantage of a blackboard system for share knowledge and effort of each agent. Agents use Reactive Model with the combination of Belief Desire Intension (BDI) Model and also use a Path Finding Algorithm for calculate shortest distance and a path for travel on the planet's surface. This approach can perform a surface exploration on a given terrain within a short period of time. Information which are gathered on the blackboard are used to make an output with detailed surface soil variance results. The developed Multi-Agent system performed well with different terrain sizes.

• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.17-32
• /
• 2006

Recently, community computing has been proposed for group formation and group decision-making. However, legacy community computing systems do not support group need identification for ad hoc group formation, which would be one of key features of ubiquitous decision support systems and services. Hence, this paper aims to provide a multi-agent based methodology to enable nomadic community computing which supports ad hoc need identification and group formation. Focusing on supporting group decision-making of relatively small sized multiple individual in a community, the methodology copes with the following three characteristics: (1) ad hoc group formation, (2) context-aware group need identification and (3) using mobile devices working in- and out-doors. NAMA-US, an RFID-based prototype system has been developed to show the feasibility of the idea proposed in this paper.

• Asia pacific journal of information systems
• /
• v.13 no.1
• /
• pp.47-72
• /
• 2003

It is important to retrieve information that a user requires on the web. The web is an open system. The amount of information is increasing rapidly. While each of information was compiled into the database piece at a single platform in the past, it is now compiled into complicated structure at a multi-platform. Restructuring the multi-platform database is needed to efficiently retrieve information. MIRAS(Meta Information Retrieval Agent System) has a multi-platform database on web. This study applies the classification of the existent site's categories to restructure the database systematically. The empirical analysis shows that the suggested method is effective for information retrieval and multi-platform database restructuring. This study helps users to save on time-cost of searching information.

• Industrial Engineering and Management Systems
• /
• v.8 no.3
• /
• pp.181-193
• /
• 2009

This paper presents a framework for controlling mobile multiple robots connected by communication networks. This framework provides novel methods to control coordinated systems using mobile agents. The combination of the mobile agent and mobile multiple robots opens a new horizon of efficient use of mobile robot resources. Instead of physical movement of multiple robots, mobile software agents can migrate from one robot to another so that they can minimize energy consumption in aggregation. The imaginary application is making "carts," such as found in large airports, intelligent. Travelers pick up carts at designated points but leave them arbitrary places. It is a considerable task to re-collect them. It is, therefore, desirable that intelligent carts (intelligent robots) draw themselves together automatically. Simple implementation may be making each cart has a designated assembly point, and when they are free, automatically return to those points. It is easy to implement, but some carts have to travel very long way back to their own assembly point, even though it is located close to some other assembly points. It consumes too much unnecessary energy so that the carts have to have expensive batteries. In order to ameliorate the situation, we employ mobile software agents to locate robots scattered in a field, e.g. an airport, and make them autonomously determine their moving behaviors by using a clustering algorithm based on the Ant Colony Optimization (ACO). ACO is the swarm intelligence-based methods, and a multi-agent system that exploit artificial stigmergy for the solution of combinatorial optimization problems. Preliminary experiments have provided a favorable result. In this paper, we focus on the implementation of the controlling mechanism of the multi-robots using the mobile agents.