• Journal of Integrative Natural Science
• /
• v.1 no.2
• /
• pp.67-75
• /
• 2008

행위자 기초 모형을 이용하여 행위자의 시장에 대한 불균일한 영향력에 대한 연구를 수행하였다. 이때 가중치를 금융시장에서 행위자 간의 공유하는 정보의 영향력의 크기로 사용하였으며 가중치의 크기와 분포가 수익의 변동에 기여하는 것을 관찰하였다. 행위자들의 가중치의 크기가 평균적으로 클수록 가격의 변동의 크기도 같이 증가함을 알 수 있었으며 가중치의 크기뿐만 아니라 가중치의 분포에 따라서도 수익의 분포가 변하게 된다. 이는 신흥시장과 성숙한 시장에서 관찰되는 분포의 차이와 관련하여 유사성을 찾아볼 수 있을 것이라는 가능성을 제공한다. 행위자의 정보의 영향력은 항상 일정하지 않고 그 영향력이 행위자의 시장 예측에 대한 적중률에 따라 변하게 된다. 이렇게 변화하는 행위자들의 정보의 영향력의 분포는 결국 소수의 큰 영향력을 갖는 행위자와 다수의 영향을 거의 끼치지 못하는 행위자들로 분포되게 된다. 그 분포는 초기의 행위자들의 영향력 분포가 어떻게 되었든 간에 충분히 시간이 흐르면 모두 멱법칙을 따르는 분포를 갖게 된다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.11a
• /
• pp.49-52
• /
• 2007

전략 게임은 여러 종류의 유닛 (Unit)이 존재한다. 각각의 유닛은 특정 유닛에 대해 강한 면모를 보이기도 하고, 또 다른 종류의 유닛에게는 약한 면모를 가지고 있다. 이를 유닛간의 상성이라고 한다. 상성은 전략적 선택을 하는데 기반이 되고, 심리전을 유발하여 보다 게임에 몰입할 수 있게 해준다. 게임 인공지능이 상성을 고려하도록 하기 위해 각각의 유닛 간에 수치화된 상성 정보가 필요하다. 그리고 생성된 수치 자료를 토대로 유닛의 행동방법을 결정할 인공 지능도 필요하게 된다. 다음 행동 및 이동을 위해 주로 사용되는 방법은 영향력 분포도(influence map)이다. 영향력 분포도는 자신과 상대방의 세력을 수치적으로 파악하는 것이다. 하지만 일반적인 형태의 영향력 분포도로는 각 유닛간의 상성을 표현하기 힘들다. 따라서 본 논문에서는 영향력 분포도를 상성에 맞게 보정할 수 있는 방법을 제시하여 인공지능이 지능적인 행동을 하도록 돕는 방법을 제안한다. 이를 길 찾기 문제에 적용하여 전략적 이동경로를 선택하는 방법을 제시하였다.

• Spatial Information Research
• /
• v.16 no.3
• /
• pp.359-372
• /
• 2008

This research estimated Pinus densiflora distribution and suggested an optimal grid size of DEM for ecologically better explaining spatial distribution and appearance pattern of Pinus densiflora and analyze appearance frequency of Pinus densiflora by topographical factors. And this work ultimately aimed at topographically analyzing spatial distribution of Pinus densiflora by comparing impact of topographical factors. The aspect was analyzed in detail, east, west, south, north, from DEM with 1m grid size, and simply, east, west, south or east, south from DEM with bigger grid size. As a result of distribution of Pinus densiflora by aspect, Pinus densiflora appeared in all aspect with smaller grid size, while distribution in specific aspect decreased with bigger grid size. The analysis of appearance frequency with 100m grid size represented spatial distribution characteristics of Pinus densiflora as visual interpretation showed. The results of comparing impact of aspect, slope, TWI proved that aspect have more impact on distribution of Pinus densiflora than slope and there was not enough a difference between impact of slope and TWI.

• Journal of Korea Multimedia Society
• /
• v.9 no.5
• /
• pp.649-657
• /
• 2006

Reinforcement Learning is a computational approach to learning whereby an agent take an action which maximize the total amount of reward it receives among possible actions within current state when interacting with a uncertain environment. Q-learning, one of the most active algorithm in Reinforcement Learning, is consist of rewards which is obtained when an agent take an action. But it has the problem with mapping real world to discrete states. When state spaces are very large, Q-learning suffers from time for learning. In constant, when the state space is reduced, many state spaces map to single state space. Because an agent only learns single action within many states, an agent takes an action monotonously. In this paper, to reduce time for learning and complement simple action, we propose the Q-learning using influence map(QIM). By using influence map and adjacent state space's learning result, an agent could choose proper action within uncertain state where an agent does not learn. When this paper compares simulation results of QIM and Q-learning, we show that QIM effects as same as Q-learning even thought QIM uses 4.6% of the Q-learning's state spaces. This is because QIM learns faster than Q-learning about 2.77 times and the state spaces which is needed to learn is reduced, so the occurred problem is complemented by the influence map.

• Journal of Digital Contents Society
• /
• v.10 no.4
• /
• pp.615-623
• /
• 2009

Games are usually composed with several units such as monsters, weapons etc.. There are often intransitive relationships between units, like the one among rock, scissors and paper. Intransitive relationships guarantee the variation of strategy choices while playing. But AIs in many games have been ignored intransitive relationships because decision making with those relationships is complex to model. This paper suggests how to use intransitive relationships to modify influence map. With the modified influence map game AI can make a different decision to win the game. With path-finding technique, this paper shows that the modified influence map makes AI's behaviors better.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.1
• /
• pp.71-86
• /
• 2010

In life insurance, the net premium is derived based on the expected life time distribution and expected interest rate. The losses or risks of the insurer are significantly affected by the obtained net premium. Thus, in life insurance, these two factors, the life time distribution and expected interest rate, are considered as important influential factors. In this paper, we investigate the effect of these influential factors on the net premiums, management risks, and the probability of losses. Furthermore, relative influence of these factors is also studied.

• Journal of the Korean Magnetics Society
• /
• v.25 no.6
• /
• pp.203-207
• /
• 2015

This paper introduces a probabilistic analysis method in order to analyze the effect of manufacturing tolerance on induction motor performance occurring in massive production. The univariate dimension reduction method is adapted to predict probabilistic characteristics of a performance function due to certain probabilistic distributions of design variables. Moreover, the sensitivity information on mean and variance of the performance function is estimated, and then the effect of randomness of individual design variables on the probability performance function is analyzed. The effectiveness and accuracy of the method is investigated with a mathematical model and an induction motor.

• Journal of Korea Game Society
• /
• v.17 no.4
• /
• pp.109-116
• /
• 2017

It takes quite amount of time to study a board game because there are many game characters and many state spaces are exist for board games. Therefore, game must do learning long. But, there is weakness with reinforcement learning. On Learning early, the learning speed becomes slow. If there were equal result that both are considered to be best ones during the course of learning stage, Heuristic which utilizes learning of problem area of Jul-Gonu was used to improve the speed of learning. To compare a normal character to an improved one, a board game was created, and then they fought against each other. As a result, improved character's ability was improved on learning speed.

• Korea Trade Review
• /
• v.41 no.5
• /
• pp.113-138
• /
• 2016

This study analyzed the impact when Korea expands Chinese Renminbi(RMB) as invoicing currency on the trade to China using Monte-Carlo simulation. Primarily, we analyzed the impact on the balance of Korean Won(KRW) converted from RMB in a case that simulated exchange rate(Korean won to Chinese Renminbi) and realized historically identical probability distribution but in different stochastic process. In addition, we developed the simulation of the case where the volatility of RMB to KRW exchange rate abnormally expanded. The major results found in this study are as follows. First, in the case where RMB exchange rate simulated in identical probability distribution but in the different stochastic process, no matter how much RMB was utilized as invoicing currency, expansion of the RMB exchange rate and exchange rate volatility operated as positive mechanism to increase the KRW converted balance. Secondly, while the expansion of US dollar exchange rate volatility positively influences the balance on average, it caused a polarization of balance, which makes under-average-balance lower and over-average-balance higher. On the contrary, the expansion of RMB exchange rate volatility even shows a similar mechanism but the impact is more moderate than USD exchange rate volatility. Thirdly, as RMB exchange rate volatility expanded, the balance of translated invoicing currency (RMB) declined, whilst the negative impact of RMB exchange rate volatility on balance of translated invoicing currency(RMB) showed diminishing effect. Lastly, the influence of RMB's exchange rate volatility through RMB usage ratio trends similar to bull spread strategy, which is a combination of call option with put option. Therefore, since RMB usage in invoicing currency could spawn a hedging effect, corporations might utilize RMB as a strategic device for maximizing profits.

• Journal of the Korean Society for information Management
• /
• v.18 no.1
• /
• pp.85-104
• /
• 2001

The purpose of the study is to investigate the international standings of Korean medical journals and to provide a concrete journal evaluation data. The SciSearch database was searched for the 7,779 papers published in 8 Korean medical journals between 1989 and 1998. The frequency of citation to each journal was measured for each year, and the journal impact factor was calculated from 1991 to 1999 exactly as it is calculated in the Jour~~lnl Citation Reports (JCR). The Korean-language journals were rarely cited in SCI, even though the journals were of the medical areas where Korean researchers published a great deal of SCI papers. The number of citations to English-language journals started to grow from 1994 and increased rapidly from 1997. The citation interval is getting shorter, resulting in the drastic increases in the impact factors in the recent years. The nine-year averages of impact factors of the journals were distributed between 0.002 and 0.126. The international standings of Korean medical journals are very poor. When the impact factors were compared to those of SCI journals in the same subject category, Korean-language journals fell below the last SCI journals ranked by the impact factor, and the English-language journals were at the bottom among the SCI journals. The impact factors of 3 English-language Korean medical journals were about 2.0 when they became SCI journals in the late 1990s.