• Journal of the Korea Society of Computer and Information
• /
• v.27 no.1
• /
• pp.9-21
• /
• 2022

Mighty is an imperfect information game, similar to the internationally popular four-player card game Bridge, but more complex in terms of game rules and operation. An environment for exploring and analyzing the strategy of the Mighty Game is required, but compared to the development of many simulators for strategy analysis of other card games such as Bridge, there is no analysis tool for the Mighty Game. Even the definition and understanding of the Mighty game at the academic level is lacking. To solve these problems, this paper systematically defined the procedures and rules of the Mighty Game. And based on this definition, we implemented a simulator that can learn Mighty game and analyze various strategies. For the usability and accessibility of the service, the simulator was developed with JavaScript, and various analysis functions are provided in the web environment. Lastly, comparative analysis with other trick-taking games dealt with in the related research domain showed that the Mighty game has its value as an incomplete information game and that there are many characteristics that make it easy to apply AI-based learning methods.

• Annual Conference of KIPS
• /
• 2018.10a
• /
• pp.708-710
• /
• 2018

강화학습은 게임의 인공지능을 대체할 수 있는 수단이지만 불완전한 게임에서 학습하기 힘들다. 학습하기 복잡한 불완전안 카드게임에서 휴리스틱한 전략을 만들고 비슷한 상태끼리 묶으면서 학습의 복잡성을 낮추었다. 인공신경망 없이 Q-러닝만으로 게임을 5만판을 통해서 상태에 따른 전략을 학습하였다. 그 결과 동일한 전략만을 사용하는 대결보다 승률이 높게 나왔고, 다양한 상태에서 다른 전략을 선택하는 것을 관찰하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.7
• /
• pp.1299-1305
• /
• 2017

In this paper, we consider a beamforming game between the transmitters in a two-user multiple-input single-output interference channel using limited feedback and investigate how each transmitter is able to find a modified strategy from the quantized channel state information (CSI). In the beamforming game, each of the transmitters (i.e., a player) tries to maximize the achievable rate (i.e., a payoff function) via a proper beamforming strategy. In our case, each transmitter's beamforming strategy is represented by a linear combining factor between the maximum ratio transmission (MRT) and the zero forcing (ZF) beamforming vectors, which is the Pareto optimal achieving strategy. With the quantized CSI, the transmitters' strategies may not be valid because of the quantization errors. We propose a modified solution, which takes into account the effects of the quantization errors.

• Journal of Korea Game Society
• /
• v.16 no.6
• /
• pp.79-90
• /
• 2016

Recently, many researchers have paid attention to the improved generation of AI agent in the area of game industry. Monte-Carlo Tree Search(MCTS) is one of the algorithms to search an optimal solution through random search with perfect information, and it is suitable for the purpose of calculating an approximate value to the solution of an equation which cannot be expressed explicitly. Games in Trading Card Game(TCG) genre such as the heartstone has imperfect information because the cards and play of an opponent are not predictable. In this study, MCTS is suggested in imperfect information card games so as to generate AI agents. In addition, the practicality of MCTS algorithm is verified by applying to heartstone game which is currently used.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.653-656
• /
• 2020

최근 새롭게 만들어진 게임 장르 중의 하나인 오토배틀러 장르는 인기가 뜨겁게 끓어오른 만큼 식는 속도 역시 빠르게 진행되었다. 침체된 오토배틀러 장르의 현 상황에서 출시 초기 유저들을 끌어들였던 오토배틀러 장르만의 특징과 차별화 된 재미요소, 흥행의 이유를 되새기고 불완전한 게임 시스템을 보완하여 유저들에게 차별화된 재미를 전달하기 위한 새로운 발전 방향에 대해 분석하고, 유저들이 해당 장르의 게임을 플레이 하면서 느낀 오토배틀러 만의 특징과 재미요소를 설문조사를 통해 파악하여 향후 오토배틀러 장르의 발전을 위하여 새로운 방향을 제시한다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.09a
• /
• pp.6-14
• /
• 1995

기업경영을 공학적으로 추상하려는 노력은 정보시스템의 개발과정에서 자연 스럽게 시작되었다. 그러나, 정보시스템 구축을 주목적으로 한 공학적 추상 은 실제 업무와 커다란 격차를 가지게 마련이다. 불완전한 업무분석이 정보 시스템 유지보수 비용의 주범이라는 지적은 이러한 주장을 뒷받침하고 있다. 기업경영에 대한 공학적 추상을 도구가 아닌 당위론적인 목적으로 시도하려 는 것은 기업경영의 내부구조를 정립하려는 시도와 동일한 것으로 이해해야 할 것이다. 본 논문은 기업경영에 대한 공학적 추상을 위한 구체적인 방법론 을 제시하고 있다. 객체지향 패러다임이 추상의 방법론으로 적용되었다.

• Journal of Digital Convergence
• /
• v.8 no.3
• /
• pp.1-17
• /
• 2010

The purpose of this study was to set up the proper method of corporal punishment. The analysis was based on game theory. Major conclusion were as follows: First, the principle of maximum utility and a rational humane of students were assumed. Second, in the game situation of imperfect information, the actions deserved punishment were done by students and punishment was not done by teachers. Third, in game situation of incomplete information, when the probability of punishment is over 50%, ill actions were done by students. Forth, in game situation of complete information, the actions deserved punishment were not done by students, punishment didn't have to be done by teachers. Fifth, in game situation of complete information, when the written agreement for punishment of teacher-students was carried out, the student's school life satisfaction measurement was improved significantly.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.19-21
• /
• 2000

This paper presents a game theory application for analyzing power transactions and market design in a deregulated energy marketplace such as PoolCo. The conventional least-cost approaches for the generation resource schedule can not exactly handle recent real-world situations. A systematic tool using game theory for the market participants is presented such that it determines the net profits through the optimal bidding strategies including the strategies for the bidding prices and bidding generations. We treat this power transaction game as incomplete information one, which means each market participants does not know other's cost function. And the demand elasticity of the energy price is considered for the realistic modeling of the deregulated marketplace.

• Environmental and Resource Economics Review
• /
• v.29 no.1
• /
• pp.47-89
• /
• 2020

We introduce a simple theoretical model to analyze the welfare impact of a competition policy in the natural gas market in South Korea. An incumbent monopolistic firm currently owns essential facilities, but the competition policy mandates that the firm provide open access to any entrant firm, charging an access fee. When no regulation is imposed on the fee pricing, this policy increases social welfare as well as the profit of the incumbent firm. When the pricing is regulated, however, social welfare depends on whether there is information asymmetry between the government and the firm regarding the operating cost of the facilities. If the government has complete information, social welfare can be maximized by choosing the optimal prices. Otherwise, the government has to set the prices based on the information that the firm delivers. We formulate a Bayesian game to analyze this case and identify a set of perfect Bayesian equilibria to compare social welfare.