• Journal of KIISE:Software and Applications
• /
• 제35권3호
• /
• pp.179-188
• /
• 2008

An important issue in multiagent reinforcement learning is how an agent should team its optimal policy through trial-and-error interactions in a dynamic environment where there exist other agents able to influence its own performance. Most previous works for multiagent reinforcement teaming tend to apply single-agent reinforcement learning techniques without any extensions or are based upon some unrealistic assumptions even though they build and use explicit models of other agents. In this paper, basic concepts that constitute the common foundation of multiagent reinforcement learning techniques are first formulated, and then, based on these concepts, previous works are compared in terms of characteristics and limitations. After that, a policy model of the opponent agent and a new multiagent reinforcement learning method using this model are introduced. Unlike previous works, the proposed multiagent reinforcement learning method utilize a policy model instead of the Q function model of the opponent agent. Moreover, this learning method can improve learning efficiency by using a simpler one than other richer but time-consuming policy models such as Finite State Machines(FSM) and Markov chains. In this paper. the Cat and Mouse game is introduced as an adversarial multiagent environment. And effectiveness of the proposed multiagent reinforcement learning method is analyzed through experiments using this game as testbed.

• Journal of Pharmaceutical Investigation
• /
• 제28권2호
• /
• pp.87-92
• /
• 1998

Kojic acid (KA) is an antimelanogenic agent which has been widely used in cosmetics to whiten the skin color. However, it has the drawbacks of the skin irritancy and the instability against the pH, temperature, and light. In order to overcome these problems, various topical gels and multiple emulsion creams which can control the release of active ingredient, KA, were formulated employing cream bases of mineral oil with caprylic capric triglyceride and hydrophilic polymers such as chitosan, carbopol. and pluronics. Using Franz diffusion cells mounted with a synthetic cellulose membrane (MWCO 12,000), drug release characteristics of the formulations were evaluated by the HPLC assay of KA concentration in the receptor compartment of pH 7.4 phosphate buffered saline solution. Drug release from chitosan-based gels (ChitoGel) obeyed to the first order kinetics with a rapid release especially in the initial period. However, pluronic-based gels (PluGel) and carbopol-based gels (CarboGel) revealed controlled release of drug to some extent, followed by the square root-time kinetics. Moreover, the release of KA was further controlled with the W/O/W multiple emulsion creams (MultiCream), showing the apparent zero order release kinetics by virtue of dynamic ratecontrolling membrane of the oil layer. The flux $(J,\;{\mu}g/cm^2/hr)$ of ChitoGel. CarboGel. PluGel. and MultiCream in the initial period of 6hr were 73.30, 28.67. 24.04 and 7.72, respectively. On the other hand, the skin irritancy score of ChitoGel and MultiCream were observed as 2.5 and 2.3 respectively, in the rabbit skin irritation test. Although there were insignificant differences at p<0.05 between those formulations, it was possible to conclude that the W/O/W multiple emulsion creams containing KA might be a good candidate for an antimelanogenic drug delivery system due to the controlled release of acidic drug molecules.

• Journal of Information Processing Systems
• /
• 제11권3호
• /
• pp.370-388
• /
• 2015

The localization of multi-agents, such as people, animals, or robots, is a requirement to accomplish several tasks. Especially in the case of multi-robotic applications, localization is the process for determining the positions of robots and targets in an unknown environment. Many sensors like GPS, lasers, and cameras are utilized in the localization process. However, these sensors produce a large amount of computational resources to process complex algorithms, because the process requires environmental mapping. Currently, combination multi-robots or swarm robots and sensor networks, as mobile sensor nodes have been widely available in indoor and outdoor environments. They allow for a type of efficient global localization that demands a relatively low amount of computational resources and for the independence of specific environmental features. However, the inherent instability in the wireless signal does not allow for it to be directly used for very accurate position estimations and making difficulty associated with conducting the localization processes of swarm robotics system. Furthermore, these swarm systems are usually highly decentralized, which makes it hard to synthesize and access global maps, it can be decrease its flexibility. In this paper, a simple pyramid RAM-based Neural Network architecture is proposed to improve the localization process of mobile sensor nodes in indoor environments. Our approach uses the capabilities of learning and generalization to reduce the effect of incorrect information and increases the accuracy of the agent's position. The results show that by using simple pyramid RAM-base Neural Network approach, produces low computational resources, a fast response for processing every changing in environmental situation and mobile sensor nodes have the ability to finish several tasks especially in localization processes in real time.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• 제5권2호
• /
• pp.133-143
• /
• 2007

This study was performed to evaluate the co- and sequential separation of Tc, Np and U from the simulated multi-component HLW solution by a TBP (tributyl phosphate)-TOA (tri- octyl amine)/NDD $(n-dodecane)-HNO_3$ extraction system. An optimal condition of (30% TBP-0.5% TOA)/NDD-1 M $HNO_3$ was selected by taking account of a prevention of the 3rd phase and effects of concentration of TBP, TOA and nitric acid on the co-extraction of Tc, Np and U. In that condition, the extraction yields were 81% (Tc), 85% (Np), less than 9% (Am and RE elements), about 8% (Pd), and less than 5% (other elements) so that the system developed for the co-extraction of Tc, Np and U was proved to be available. For that, however, more than 99% of Zr was found to be pre-removed. The co-extracted Tc, Np and U were sequentially separated in order of Tc(stripping agent : 5 M $HNO_3$)${\rightarrow}Np$ by reductive stripping (reductive-stripping agent : 0.1 M AHA)${\rightarrow}U$ (stripping agent : 0.01 M $HNO_3$), and then their separation factors were evaluated. At these conditions, 95% of Tc, 98% of Np and 99% of U could be recovered in each step.

• Journal of KIISE:Computing Practices and Letters
• /
• 제7권6호
• /
• pp.641-652
• /
• 2001

Most of current internet auction systems are single auction server program that partly automates the function of the conventional auction house. So these systems do not provide sufficient independence, distribution and parallelism between the functions of the conventional auction house. In these suction systems, only some of suction management facilities are automated by the server program, but still users need to execute a repetitive task to monitor the dynamic progress of a certain auction, decide the proper bid price, and submit the bid. Another problem is that they support only the single auctions such as English suction and Dutch auction, but they do not support the double auctions that are superior to the single auction in terms of speed, efficiency, and the fair distribution of profit. In this paper, we present are design and implement of an agent-based continuous double auction system, called CoDABot, in order to overcome the limitations of current auction systems. CoDABot supports the continuous double auction, provides various bidding agents for users to select, and has been implemented with a multi-agent system to realize more independent, and parallel subsystems.

• Journal of the Korean Applied Science and Technology
• /
• 제19권1호
• /
• pp.10-18
• /
• 2002

In this study, it should be mentioned that Lipid-LCG can be prepared with the main compound of hydrogenated lecithin in oil-in water emulsion. The results of its physical property and stability are as follows. First, the best suitable compositions of Lipid-LCG are made from 4.0wt% of the hydrogenated lecithin, 4.0wt% of cetostearyl alcohol as emulsifier and gelling agent, 3.0wt% of butylene glycol and 2.0wt% glycerin as moisturizers, 3.0wt% of cyclomethicone, 3.0wt% of isononyl-isononanoate, 3.0wt% of capric/caprylic triglycerides, 3.0wt% of macadamia oil as emollients. Second, As the optimum conditions to form Lipid-LCG, which figured out 6.0 ${\pm}$ 1.0 for pH level, 32kg/mm, min for hardness to make a .essence to be formed the ternary phase of liquid crystal(multi-lamellar type). Third, as the analytical result of this system, it obtained that particle size is $1{\sim}8{\mu}m$ level, and is certified with it at 400 and 1,000 magnifications by microscope. The stability of Lipid-LCG is very stable on condition of a low temperature ($4^{\circ}C$), a room temperature ($25^{\circ}C$) and a high temperature ($40^{\circ}C$), which is not to be split in for a long time(for 3-month). We produced our own moisturizing essence, which has a good affinity to skin by means of this system.

• The KIPS Transactions:PartB
• /
• 제10B권7호
• /
• pp.751-756
• /
• 2003

Ant Colony Optimization (ACO) is new meta heuristics method to solve hard combinatorial optimization problem. It is a population based approach that uses exploitation of positive feedback as well as greedy search. It was firstly proposed for tackling the well known Traveling Salesman Problem (TSP) . In this paper, we introduce Multi Colony Ant Model that achieve positive interaction and negative interaction through Intensification and Diversification to improve original ACS performance. This algorithm is a method to solve problem through interaction between ACS groups that consist of some agent colonies to solve TSP problem. In this paper, we apply this proposed method to TSP problem and evaluates previous method and comparison for the performance and we wish to certify that qualitative level of problem solution is excellent.

• Food Science of Animal Resources
• /
• 제44권4호
• /
• pp.873-884
• /
• 2024

Flunixin is a veterinary nonsteroidal anti-inflammatory agent whose residues have been investigated in their original form within tissues such as muscle and liver. However, flunixin remains in milk as a metabolite, and 5-hydroxy flunixin has been used as the primary marker for its surveillance. This study aimed to develop a quantitative method for detecting flunixin and 5-hydroxy flunixin in milk and to strengthen the monitoring system by applying to other livestock and fishery products. Two different methods were compared, and the target compounds were extracted from milk using an organic solvent, purified with C₁₈, concentrated, and reconstituted using a methanol-based solvent. Following filtering, the final sample was analyzed using liquid chromatography-tandem mass spectrometry. Method 1 is environmentally friendly due to the low use of reagents and is based on a multi-residue, multi-class analysis method approved by the Ministry of Food and Drug Safety. The accuracy and precision of both methods were 84.6%-115% and 0.7%-9.3%, respectively. Owing to the low matrix effect in milk and its convenience, Method 1 was evaluated for other matrices (beef, chicken, egg, flatfish, and shrimp) and its recovery and coefficient of variation are sufficient according to the Codex criteria (CAC/GL 71-2009). The limits of detection and quantification were 2-8 and 5-27 ㎍/kg for flunixin and 2-10 and 6-33 ㎍/kg for 5-hydroxy flunixin, respectively. This study can be used as a monitoring method for a positive list system that regulates veterinary drug residues for all livestock and fisheries products.

• Proceedings of the Korean Information Science Society Conference
• /
• 한국정보과학회 2008년도 한국컴퓨터종합학술대회논문집 Vol.35 No.1 (D)
• /
• pp.307-312
• /
• 2008

최근 전기 전력 도메인의 IT화 (전력 IT) 가 본격적으로 진행되면서 전기 데이터를 취득하기 위한 기기들 뿐만 아니라 전력 운영 환경이 변화를 도모하고 있다 본 논문에서는 이런 전력 IT 의 일환으로 개발된 지능형 전력 정보 모니터링 시스템 (K-WAMS)을 소개하고자 한다. 이 시스템은 이기종 분산 환경에서 전력 정보를 실시간으로 전송 받아 현재 전력계통의 상황을 직관적으로 판단할 수 있는 정보를 제공하여 광역 정전을 예방하고자 하는 조기경보시스템 (Early Warning System) 이다. 또한, 전력계통의 안정성 여부를 판단하는 지능형 에이전트의 구현을 통해 광역 정전을 예견하는 지수를 제공하고, 실시간 데이터 처리 및 이력 데이터 저장 에이전트들을 통해 현재 취득되고 있는 현장 데이터 모니터링 기능을 제공한다. K-WAMS 는 실시간으로 전송 처리되는 고속 대용량 데이터들의 처리 지연, 동적 환경으로 인한 네트워크의 부하문제를 해결하기 위해서 IPC Shared Memory 기법을 이용하였고, 동적 프로세스들의 작업 스케줄링을 관리하기 위한 IPC Message Queue 기법을 이용하였다. 또한 논리적 기능 기반으로 설계된 현재 중앙 급전소(GCC) 단위의 설계 모델은 향후 지역 급전소(RCC) 단위로 설치 될 경우 재사용 되어 개발 생산성을 향상시킬 것으로 예상된다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• 제13권4호
• /
• pp.3-11
• /
• 2003

It is very important to detect and remove original sources of various attacks through networks. One of the effective method to detect the sources is traceback systems. In this paper, we design and implement an agent-based traceback system that does not require the reaction of routers and administrators and does not need numerous log data. In the design, we introduce a traceback server and traceback agents in each network Using sniffing and spoofing, the server transmits a packet with a specific message. The agents detect the packet and provide the information for the server to trace back the original source.