• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.6
• /
• pp.69-80
• /
• 2004

This paper proposes a method to learn action patterns of opponent characters for intelligent characters. For learning action patterns, intelligent characters learn the past actions as well as the current actions of opponent characters. Therefore, intelligent characters react more properly than ones without the knowledge on action patterns. In addition, this paper proposes a method to learn moving actions whose fitness is hard to evaluate. To evaluate the performance of the proposed algorithm, we experiment with four repeated action patterns in a game similar to real games. The results show that intelligent characters learn the optimal actions for action patterns and react properly against to random action opponent characters. The proposed method can be applied to various games in which characters confront each other, e.g. massively multiple of line games.

• The Korean Journal of Applied Statistics
• /
• v.21 no.3
• /
• pp.441-453
• /
• 2008

We analyzed Korean professional basketball and baseball players salary under the assumption that it depends on the personal records and contribution to the team in the previous year. We extensively used data visualization tools to check the relationship among the variables, to find outliers and to do model diagnostics. We used multiple linear regression and regression tree to fit the model and used cross-validation to find an optimal model. We check the relationship between variables carefully and chose a set of variables for the stepwise regression instead of using all variables. We found that points per game, number of assists, number of free throw successes, career are important variables for the basketball players. For the baseball pitchers, career, number of strike-outs per 9 innings, ERA, number of homeruns are important variables. For the baseball hitters, career, number of hits, FA are important variables.

• Environmental and Resource Economics Review
• /
• v.25 no.2
• /
• pp.255-276
• /
• 2016

This paper is designed to investigate the dynamic structure of optimal tariff imposed by the oil importing country from OPEC when OPEC increases the oil reserves through exploration and development. For this purpose, we used a Stckelberg differential game modeling approach, and tried to analyze two cases such that the one is the extraction cost is dependent on the oil reserves and the other is the extraction cost is independent of it. On the basis of this analysis, we propose that the importing contry has to impose dynamically inconsistent tariffs if the extraction cost is dependent on the oil reserves. Otherwise, she should impose dynamically consistent tariffs announced at initial time. In addition, we found that whether or not the exporting country uses some portion of oil produced for domestic consumption does not affect our policy proposition stated above.

• Journal of Korean Society of Transportation
• /
• v.27 no.6
• /
• pp.147-156
• /
• 2009

We present a methodology for modeling and solving the transit frequency design problem with variable demand. The problem is described as a bi-level model based on a non-cooperative Stackelberg game. The upper-level operator problem is formulated as a non-linear optimization model to minimize net cost, which includes operating cost, travel cost and revenue, with fleet size and frequency constraints. The lower-level user problem is formulated as a capacity-constrained stochastic user equilibrium assignment model with variable demand, considering transfer delay between transit lines. An efficient algorithm is also presented for solving the proposed model. The upper-level model is solved by a gradient projection method, and the lower-level model is solved by an existing iterative balancing method. An application of the proposed model and algorithm is presented using a small test network. The results of this application show that the proposed algorithm converges well to an optimal point. The methodology of this study is expected to contribute to form a theoretical basis for diagnosing the problems of current transit systems and for improving its operational efficiency to increase the demand as well as the level of service.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.4072-4089
• /
• 2018

This paper introduces a scheme for optimizing the material properties of mass-spring systems of different resolutions to provide coherent behavior for reduced level-of-detail in MSS(Mass-Spring System) meshes. The global optimal material coefficients are derived to match the behavior of provided reference mesh. The proposed method also gives us insight into levels of reduction that we can achieve in the systematic behavioral coherency among the different resolution of MSS meshes. We obtain visually acceptable coherent behaviors for cloth models based on our proposed error metric and identify that this method can significantly reduce the resolution levels of simulated objects. In addition, we have confirmed coherent behaviors with different resolutions through various experimental validation tests. We analyzed spring force estimations through triangular Barycentric coordinates based from the reference MSS that uses a Gaussian kernel based distribution. Experimental results show that the displacement difference ratio of the node positions is less than 10% even if the number of nodes of $MSS^{sim}$ decreases by more than 50% compared with $MSS^{ref}$. Therefore, we believe that it can be applied to various fields that are requiring the real-time simulation technology such as VR, AR, surgical simulation, mobile game, and numerous other application domains.

• KIPS Transactions on Computer and Communication Systems
• /
• v.1 no.3
• /
• pp.133-142
• /
• 2012

Cournot model is one of representative models among many game theoretic approaches available for analyzing competitive market models. Recent years have witnessed various kinds of attempts to model competitive electricity markets using the Cournot model. Cournot model is appropriate for oligopoly market which is one characteristic of electric power industry requiring huge amount of capital investment. When we use Cournot model for the application to electricity market, it is prerequisite to assume the downward sloping demand curve in the right direction. Generators in oligopoly market could try to maximize their profit by exercising the market power like physical or economic withholding. However advanced electricity markets also have demand side bidding which makes it possible for the demand to respond to the high market price by reducing their consumption. Considering this kind of demand reaction, Generators couldn't abuse their market power. Instead, they try to find out an equilibrium point which is optimal for both sides, generators and demand. This paper suggest a quantitative analysis between market variables based on econometrics for estimating demand responses in smart grid environment.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.41 no.2
• /
• pp.145-152
• /
• 2017

Mixing of molecular grains having different characteristics is very important in many industries such as the food and pharmaceutical industries. With the development of computer simulations, it is common practice to find the optimal mixing conditions through a simulation before the actual mixing task to estimate the proper level of mixing. Accordingly, there has been an increasing need for a mixing index to measure the mix of particles in the simulation process. Mixing indices, which have been widely used so far, can largely be classified into two types: first is the statistical-based mixing index, which is prepared using the sampling method, and the second is the mixing index that is prepared using all the particles. In this paper, we calculated mixing indices in different ways for the data in the course of mixing the particles using the DEM simulation. Additionally, we compared the performance, advantages, and disadvantages of each mixing index. Therefore, I propose a standard that can be used to select an appropriate mixing index.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.2
• /
• pp.31-40
• /
• 2012

Iris recognition is a biometric technology which can identify a person using the iris pattern. It is important for the iris recognition system to extract the feature which is invariant to changes in iris patterns. Those changes can be occurred by the influence of lights, changes in the size of the pupil, and head tilting. In this paper, we propose a novel method based on Zernike Moment which is robust to rotations of iris patterns. we utilized a selection of Zernike moments for the fast and effective recognition by selecting global optimum moments and local optimum moments for optimal matching of each iris class. The proposed method enables high-speed feature extraction and feature comparison because it requires no additional processing to obtain the rotation invariance, and shows comparable performance to the well-known previous methods.

• Korean Journal of Optics and Photonics
• /
• v.29 no.6
• /
• pp.253-261
• /
• 2018

In this paper, we study a method of designing a reflector for LED light sources with an asymmetric light distribution. In a sports game, lighting with a symmetric distribution makes the athlete and spectators look directly at the light source, so it can cause glare. We derive the optimal tilt angle and design a reflector with asymmetric light distribution to solve these problems. Afterward, performance is analyzed according to the tennis-court lighting standard, and is confirmed to meet the class 1 European standard.

• Journal of Advanced Navigation Technology
• /
• v.26 no.2
• /
• pp.72-77
• /
• 2022

Among the techniques of reinforcement learning, Q-Learning means learning optimal policies by learning Q functions that perform actionsin a given state and predict future efficient expectations. Q-Learning is widely used as a basic algorithm for reinforcement learning. In this paper, we studied the effectiveness of selecting and learning efficient paths by designing policies and rewards based on Q-Learning. In addition, the results of the existing algorithm and punishment compensation policy and the proposed punishment reinforcement policy were compared by applying the same number of times of learning to the 8x8 grid environment of the Frozen Lake game. Through this comparison, it was analyzed that the Q-Learning punishment reinforcement policy proposed in this paper can significantly increase the learning speed compared to the application of conventional algorithms.