Abstract
The topic of this paper is an analysis of Werewolf game. In this game, the players are divided into the villagers’ side and the werewolves’ side. The players erasure the opposite side’s players through discussion and inference. In this paper, execution probabilities and attack probabilities are predicted using machine learning. Five-player games are adopted as the fundamental setting. Game logs were collected from an online chat-style Werewolf game server and analyzed as training data using logistic regression. The authors then analyzed the test data (game logs) with the calculated partial regression coefficients. For both execution and attack probabilities, predictions were appropriate when the self-proclaimed seer appeared. Since the appearance of the seer(s) is an objective piece of information about the situation, it is properly reflected in the prediction. In these predictions, the probabilities indicate which players are certainly executed/attacked otherwise which players are certain to survive. When no seers appeared, because there was little objective information, the execution/attack probabilities of each player changed little. In these situations, since the players can’t decide who to execute/attack, the predictions are not irrational. In short, when the seer(s) (who provide objective information to the field) appear, the predictions tend to be appropriate. In addition, the predictions in this paper are stable with not much (50–80) training data, which is one of the characteristics of this method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Braverman, M., Etesami, O., Mossel, E.: Mafia: a theoretical study of players and coalitions in a partial information environment. Ann. Appl. Probab. 18(3), 825–846 (2008)
Yao, E.: A theoretical study of Mafia games. arXiv:0804.0071 (2008)
Migdal, P.: A mathematical model of the Mafia game. arXiv:1009.1031 (2013)
Nishino, J.: A natural winning percentage of the werewolf game. IPSJ Technical Report, vol. 2015-GI-33, no. 18 (2015). (in Japanese)
Miyata, H., Arakawa, T.: Mathematical models of Werewolf with execution probabilities and attack probabilities. In: Game Programming Workshop, vol. 2019, pp. 161–164 (2019). (in Japanese)
Inaba, M., Ohata, N., Takahashi, K., Toriumi, F.: You will be executed if you keep chatting —development of tag sets for utterances in the werewolf game and analysis of action and winning or losing using the tags. J. IPSJ 57(11), 2392–2402 (2016). (in Japanese)
Inaba, M., Osawa, H., Katagami, D., Shinoda, K., Toriumi, F.: Analysis of werewolf game based on discussion structure. In: Game Programming Workshop, vol. 2014, pp. 61–66 (2014). (inJapanese)
Shimizu, D., Hasebe, K.: Game theoretic analysis of best utterance in Werewolf game. In: The 37th JSSST Annual Conference, 55-L (2020). (in Japanese)
Inaba, M., Toriumi, F., Takahashi, K.: The statistical analysis of Werewolf game data. In: Game Programming Workshop, vol. 2012, pp. 144–147 (2012). (in Japanese)
Sonoda, A., Toriumi, F.: Analysis of trust in “are you a werewolf?” In: The 31st Annual Conference of JSAI, 2H1-1in1 (2017). (in Japanese)
Konno, N., Otsuki, T.: Improvement of team estimation in AIWolf player using machine learning. In: Entertainment Computing Symposium, vol. 2017, pp. 64–69 (2017). (in Japanese)
Kimura, Y., Ito, T.: Improvement of role estimation using machine learning in Werewolf AI. In: The 32nd Annual Conference of JSAI, 1H1-OS-13a-03 (2018). (in Japanese)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Saito, R., Ushida, K., Arakawa, T. (2024). Prediction of Execution Probabilities and Attack Probabilities of Werewolf Game Using Machine Learning of Game Logs. In: Lee, CY., Lin, CL., Chang, HT. (eds) Technologies and Applications of Artificial Intelligence. TAAI 2023. Communications in Computer and Information Science, vol 2075. Springer, Singapore. https://doi.org/10.1007/978-981-97-1714-9_13
Download citation
DOI: https://doi.org/10.1007/978-981-97-1714-9_13
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-97-1713-2
Online ISBN: 978-981-97-1714-9
eBook Packages: Computer ScienceComputer Science (R0)