• Journal of Korea Game Society
• /
• v.9 no.2
• /
• pp.125-133
• /
• 2009

Game artificial intelligence is usually used to provide intelligent and adjusted game environment for user. Previously, it was used for Non-player character(NPC) playing a role of a company or an enemy through collecting and analyzing a user's behaviour. However, it was just mimicking the user's behavior. This paper introduces a method to change game environment by analyzing a user's game behavior. Game behavior data has been used to understand user's game preference. Also, the user's preference was used to provide more active game environment by reflecting decision of geographical features, items and distribution of NPC. For experiment of the suggested method, we utilized a real 2D action game and confirmed the game environment which changing properly according to the user's game play.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.4
• /
• pp.785-794
• /
• 2009

We studied on a technic to improve and implement the virtual environment grid information in the war-game model. The virtual environment data are categorized into ground, sea, and air, and separatedly managed and operated. The data are implemented and managed in the geographical information and topography type information. The geographical information, including grid location coordinate and sea level, are managed in computer files as a grid type. We improved the representation of the sea level by mapping the sea level value of the grid information file into GCV and suggested the dynamic virtual environment implementation technic for the region with the large range of tide. In the experiments, we could know that it is possible to adopt the dynamic virtual environment to the war-game model.

• Journal of Korea Game Society
• /
• v.9 no.4
• /
• pp.3-10
• /
• 2009

Recently, various ways are being explored for enhancing the fun of computer games and lengthening the life cycle of them. This paper proposes the more active game environment by reflecting the decision that is made by collecting and analyzing user's behavioral data. The game results and traveling information of user is using for dynamic map creation. For experiment of the proposed method, we created games using the XNA Framework and confirmed the game environment which changing properly according to the user's game play.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.52 no.9
• /
• pp.542-549
• /
• 2003

In this paper, a novel approach to generator maintenance scheduling strategy in competitive electricity markets based on non-cooperative dynamic game theory is presented. The main contribution of this study can be considered to develop a game-theoretic framework for analyzing strategic behaviors of generating companies (Gencos) from the standpoints of the generator maintenance-scheduling problem (GMP) game. To obtain the equilibrium solution for the GMP game, the GMP problem is formulated as a dynamic non-cooperative game with complete information. In the proposed game, the players correspond to the profit-maximizing individual Gencos, and the payoff of each player is defined as the profits from the energy market. The optimal maintenance schedule is defined by subgame perfect equilibrium of the game. Numerical results for two-Genco system by both proposed method and conventional one are used to demonstrate that 1) the proposed framework can be successfully applied in analyzing the strategic behaviors of each Genco in changed markets and 2) both methods show considerably different results in terms of market stability or system reliability. The result indicates that generator maintenance scheduling strategy is one of the crucial strategic decision-makings whereby Gencos can maximize their profits in a competitive market environment.

• Journal of Korea Game Society
• /
• v.15 no.1
• /
• pp.115-124
• /
• 2015

This research analyzed the case of Ncsoft to study the organizational learning, exploitation and exploration that create dynamic capability in hypercompetitive environment. First of all, we demonstrated the activities of exploitation and exploration in Ncsoft according to the life cycle of online game industry. An exploitation related to routine, learning and fit with existing environment brings about incremental innovation. In contrast, an exploration associated with non-learning, flexibility with changing environment results in radical innovation. We examined them based on the life cycle of its various game services. NCsoft that built the leading position in online game industry focused the exploitation activities at the stage of beginning period and growth, whereas NCsoft has increased the activities of exploration at period of mature. In addition, the firm conducts an exploration for its brand new game services and R&D. Conversely, An exploitation is conducted for sustainable updating of patch service and marketing and system building. The result implies that online game companies create sustainable competitive advantage using the balance between exploitation and exploration.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.149.5-149
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent´s dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless ...

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.10
• /
• pp.3867-3886
• /
• 2015

This article investigates the problem of opportunistic spectrum access in dynamic environment, in which the signal-to-noise ratio (SNR) is time-varying. Different from existing work on continuous feedback, we consider more practical scenarios in which the transmitter receives an Acknowledgment (ACK) if the received SNR is larger than the required threshold, and otherwise a Non-Acknowledgment (NACK). That is, the feedback is discrete. Several applications with different threshold values are also considered in this work. The channel selection problem is formulated as a non-cooperative game, and subsequently it is proved to be a potential game, which has at least one pure strategy Nash equilibrium. Following this, a multi-agent Q-learning algorithm is proposed to converge to Nash equilibria of the game. Furthermore, opportunistic spectrum access with multiple discrete feedbacks is also investigated. Finally, the simulation results verify that the proposed multi-agent Q-learning algorithm is applicable to both situations with binary feedback and multiple discrete feedbacks.

• Journal of Korea Game Society
• /
• v.4 no.3
• /
• pp.55-64
• /
• 2004

A dynamic graph is suitable for representing and managing dynamic changable obstacles or terrain information in 2D/3D games such as RPG and Strategy Simulation Games. We propose a dynamic hierarchical graph model with fixed level to perform a quick A* path finding. We divide a graph into subgraphs by using space classification and space model, and construct a hierarchical graph. And then we perform a quick path fading on the graph by using our dynamic graph operators. With our experiments we show that this graph model has efficient properties for finding path in a dynamic game environment.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.52 no.11
• /
• pp.668-674
• /
• 2003

The electric marketplace is in the midst of major changes designed to promote competition. No longer vertically integrated with guaranteed customers and suppliers, electric generators and distributors will have to compete to sell and buy electricity. Unit commitment (UC) in such a competitive environment is not the same as the traditional one anymore. The objective of UC is not to minimize production cost as before but to find the solution that produces a maximum profit for a generation firm. This paper presents a hi-level formulation that decomposes the UC game into a generation-decision game (first level game) and a state(on/off)-decision game (second level game). Derivation that the first-level game has a pure Cournot Nash equilibrium(NE) helps to solve the second-level game. In case of having a mixed NE in the second-level game, this paper chooses a pure strategy having maximum probability in the mixed strategy in order to obviate the probabilistic on/off state which may be infeasible. Simulation results shows that proposed method gives the adequate UC solutions corresponding to a NE.

• Journal of Korea Game Society
• /
• v.6 no.1
• /
• pp.3-10
• /
• 2006

In computer games, the level design and balance of characters are the key features for developing interesting games. Level designers make decision to change the parameters and opponent behaviors in order to avoid the player getting extremely frustrated with the improper level. Generally, opponent behavior is defined by static script, this causes the games to have static difficulty level and static environment. Therefore, it is difficult to keep track of the user playing interest, because a player can easily adapt to changeless repetition. In this paper, we propose a dynamic scripting method that able to maintain the level designers' intention where user enjoys the game by adjusting the opponent behavior while playing the game. The player's countermeasure pattern for dynamic level design is modeled using a Gaussian Mixture Model (GMM). The proposed method is applied to a shooting game, and the experimental results maintain the degree of interest intended by the level designer.