게임플레이어는 게임에서 수많은 적들을 만나고 싸우게 되는데, 이 때 너무나 손쉽게 이기거나 진다면 게임의 재미는 반감될 것이다. 그 반대로 너무 어렵게 이긴다거나 지는 것도 게임을 지루하게 만드는 요인으로 작용한다. 따라서 상대방과의 전투나 경쟁에서 아슬아슬하게 승리하는 긴장감을 주기 위해서는 게임의 밸런스가 잘 맞아야 한다. 그만큼 게임 밸런싱 작업은 게임의 재미와 가장 직접적으로 영향을 미치는 요소로 작용한다. 그리고 게임 밸런스만큼 중요한 것이 있는데, 그것은 플레이어에게 적절한 난이도의 상대를 계속 만나게 하는 것이다. 본 연구에서는 이러한 문제를 해결하려는 방법으로써 게임 밸런스에 딥러닝을 적용하여 지능 캐릭터가 플레이어를 통해 학습하고 스스로 플레이어의 난이도에 따라 자신의 난이도를 조절할 수 있도록 고안하였다. 이것이 활성화되면 게임 기획자나 개발자에게는 그만큼의 비용을 절약하는 동시에 플레이어에게는 항상 흥미로운 상대를 제공할 수 있는 획기적인 방법이 될 것이다.
게임플레이어는 게임에서 수많은 적들을 만나고 싸우게 되는데, 이 때 너무나 손쉽게 이기거나 진다면 게임의 재미는 반감될 것이다. 그 반대로 너무 어렵게 이긴다거나 지는 것도 게임을 지루하게 만드는 요인으로 작용한다. 따라서 상대방과의 전투나 경쟁에서 아슬아슬하게 승리하는 긴장감을 주기 위해서는 게임의 밸런스가 잘 맞아야 한다. 그만큼 게임 밸런싱 작업은 게임의 재미와 가장 직접적으로 영향을 미치는 요소로 작용한다. 그리고 게임 밸런스만큼 중요한 것이 있는데, 그것은 플레이어에게 적절한 난이도의 상대를 계속 만나게 하는 것이다. 본 연구에서는 이러한 문제를 해결하려는 방법으로써 게임 밸런스에 딥러닝을 적용하여 지능 캐릭터가 플레이어를 통해 학습하고 스스로 플레이어의 난이도에 따라 자신의 난이도를 조절할 수 있도록 고안하였다. 이것이 활성화되면 게임 기획자나 개발자에게는 그만큼의 비용을 절약하는 동시에 플레이어에게는 항상 흥미로운 상대를 제공할 수 있는 획기적인 방법이 될 것이다.
Game balance settings are crucial to game design. Game balancing must take into account a large amount of numerical values, configuration data, and the relationship between elements. Once released and served, a game - even for a balanced game - often requires calibration according to the game player...
Game balance settings are crucial to game design. Game balancing must take into account a large amount of numerical values, configuration data, and the relationship between elements. Once released and served, a game - even for a balanced game - often requires calibration according to the game player's preference. To achieve sustainability, game balance needs adjustment while allowing for small changes. In fact, from the producers' standpoint, game balance issue is a critical success factor in game production. Therefore, they often invest much time and capital in game design. However, if such a costly game cannot provide players with an appropriate level of difficulty, the game is more likely to fail. On the contrary, if the game successfully identifies the game players' propensity and performs self-balancing to provide appropriate difficulty levels, this will significantly reduce the likelihood of game failure, while at the same time increasing the lifecycle of the game. Accordingly, if a novel technology for game balancing is developed using artificial intelligence (AI) that offers personalized, intelligent, and customized service to individual game players, it would bring significant changes to the game production system.
Game balance settings are crucial to game design. Game balancing must take into account a large amount of numerical values, configuration data, and the relationship between elements. Once released and served, a game - even for a balanced game - often requires calibration according to the game player's preference. To achieve sustainability, game balance needs adjustment while allowing for small changes. In fact, from the producers' standpoint, game balance issue is a critical success factor in game production. Therefore, they often invest much time and capital in game design. However, if such a costly game cannot provide players with an appropriate level of difficulty, the game is more likely to fail. On the contrary, if the game successfully identifies the game players' propensity and performs self-balancing to provide appropriate difficulty levels, this will significantly reduce the likelihood of game failure, while at the same time increasing the lifecycle of the game. Accordingly, if a novel technology for game balancing is developed using artificial intelligence (AI) that offers personalized, intelligent, and customized service to individual game players, it would bring significant changes to the game production system.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
Such research includes analysis of monsters and their AI that are closely related to a character’s battle experience.
This is because online games may run into unexpected circumstances (breaking into a lobby for games with AI characters or attacking a player character, for example) because of interactions between players, and player experientiality is established in diverse ways for standalone and multiplayer games. Therefore, the scope of this study will be limited to the games that use AI characters.
제안 방법
In addition, the proposed method does not require a complete definition of the AI character’s behaviors and conditions that uses finite-state machine (FSM) or flowcharts; rather, the AI character would adjust the difficulty level tailored to its player through self-learning from the player’s behavior and decision making.
In the study of Shin Yong-woo and Jeong Tae-Chung, similar to this concept, an attempt was made to predict a player’s behavior by using sequential prediction and N-grams models.
There are nine most frequently used combat skills, assigned to keypad 1 through 9, providing quick access to the skills. Taking into account the system status, this study analyzes the Mage abilities, mainly with the frequently used skills. Information about the nine skills can be loaded in the Skill.
The method proposed in this study was designed to self-adjust game difficulty levels by identifying the player’s propensity (combat pattern, player character settings, etc.) across all game contents using AI characters.
This study proposes a method of game balancing using deep learning techniques. The proposed method will save time and effort in the process where a player character is first set up and accordingly balanced as intended. In addition, the proposed method does not require a complete definition of the AI character’s behaviors and conditions that uses finite-state machine (FSM) or flowcharts; rather, the AI character would adjust the difficulty level tailored to its player through self-learning from the player’s behavior and decision making.
This study proposes a method of game balancing using deep learning techniques. The proposed method will save time and effort in the process where a player character is first set up and accordingly balanced as intended.
This study will use the Weka framework that supports multilayer perceptrons and trains them with backpropagation for the purpose of full-scale neural network configuration.
이론/모형
For the configuration of an artificial neural network, this study uses “Weka”, a data mining program that allows AI characters to calculate the players’ combat patterns and determine their optimized behaviors and settings through the application of Multilayer perceptron.
성능/효과
649. In conclusion, Player 2 (coded as xin the program) is more likely to win the battle when Frost-Nova is the first skill used in combat. If Intelligent character can compute Q-Values in the same way as above and select higher values through the Monte Carlo tree search, then the optimal skill pattern can be created.
후속연구
On the other hand, programs make relatively less mistakes or have less of a factor that lowers the permanency compared to humans. In this sense, this study will deliver significant cost savings from the design sector while facilitating reinvestment in other sectors, leading to a higher success rate of new game introductions in the market. In addition, If the user’s physical ability is numerically applied in a medical game field where a game level system is used.
This study will also significantly save time and costs required for game balancing in the related industries. Much of game design costs are labor related.
참고문헌 (14)
오별. AI를 이용한 게임 밸런스 방법. 정보과학회지. 27(10). 한국정보과학회. pp. 25-28. 2009
※ AI-Helper는 부적절한 답변을 할 수 있습니다.