Battle Prediction System in StarCraft
Combined with Topographic Considerations
ChengZhen Meng, Yu Tang
(&) , ChenYao Wu, and Di Lin
University of Electronic Science and Technology of China, Shahe Campus:
No. 4, Section 2, North Jianshe Road, Chengdu 610054, Sichuan,
People’s Republic of China
yutang@uestc.edu.cn
Abstract. This paper focuses on the prediction of combat outcomes in a local
battle during a game of StarCraft. Through the analysis of the initial state of the
two armies and considering the influence of the terrain in StarCraft on the
combat effectiveness of both sides, the concept of the terrain correction factor is
introduced to establish a mathematical model. Secondly, using SparCraft is to
simulate battles and generate data sets. Finally, the maximum a posteriori
probability estimation (MAP) is used to train the previously established data set
to complete the parameter estimation of the mathematical model.
Keywords: SparCraft Á MAP Á Bayesian formula
1 Overview
With the continuous development of machine learning technology, artificial intelligence has received more and more attention from society, and people have more and
more demand for high-performance artificial intelligence. As a result, many artificial
intelligence-related competitions have emerged. Real-time strategy games (RTS),
because of their unique tactical nature and a well-defined complex confrontation system they represented, which can be divided into many interesting subproblems, have
become a hot area for testing artificial intelligence.
With regard to the field of artificial intelligence in StarCraft, there are many
directions to be studied, from pathfinding to image recognition, strategic decision
making, etc., and the core point is to judge the outcome of small-scale battles in a
game, accurately predicting the outcome of the battle will be regard as a very important
reference for AI’s subsequent tactical decisions and actions. In StarCraft, the factors
involved in a battle are mainly the size and composition of the armed forces on both
sides of the war. The main purpose is to improve the model-related parameters through
the mathematical composition of the two sides, through mathematical modeling, and
combined with Bayesian classifier algorithm, and finally achieve the purpose of more
accurate prediction of the battle results. Namely given the size and composition of
Army A and Army B, it is possible to answer more accurately: Who will win? [1].
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 122–130, 2020.
https://doi.org/10.1007/978-981-15-0187-6_14
Combined with Topographic Considerations
ChengZhen Meng, Yu Tang
(&) , ChenYao Wu, and Di Lin
University of Electronic Science and Technology of China, Shahe Campus:
No. 4, Section 2, North Jianshe Road, Chengdu 610054, Sichuan,
People’s Republic of China
yutang@uestc.edu.cn
Abstract. This paper focuses on the prediction of combat outcomes in a local
battle during a game of StarCraft. Through the analysis of the initial state of the
two armies and considering the influence of the terrain in StarCraft on the
combat effectiveness of both sides, the concept of the terrain correction factor is
introduced to establish a mathematical model. Secondly, using SparCraft is to
simulate battles and generate data sets. Finally, the maximum a posteriori
probability estimation (MAP) is used to train the previously established data set
to complete the parameter estimation of the mathematical model.
Keywords: SparCraft Á MAP Á Bayesian formula
1 Overview
With the continuous development of machine learning technology, artificial intelligence has received more and more attention from society, and people have more and
more demand for high-performance artificial intelligence. As a result, many artificial
intelligence-related competitions have emerged. Real-time strategy games (RTS),
because of their unique tactical nature and a well-defined complex confrontation system they represented, which can be divided into many interesting subproblems, have
become a hot area for testing artificial intelligence.
With regard to the field of artificial intelligence in StarCraft, there are many
directions to be studied, from pathfinding to image recognition, strategic decision
making, etc., and the core point is to judge the outcome of small-scale battles in a
game, accurately predicting the outcome of the battle will be regard as a very important
reference for AI’s subsequent tactical decisions and actions. In StarCraft, the factors
involved in a battle are mainly the size and composition of the armed forces on both
sides of the war. The main purpose is to improve the model-related parameters through
the mathematical composition of the two sides, through mathematical modeling, and
combined with Bayesian classifier algorithm, and finally achieve the purpose of more
accurate prediction of the battle results. Namely given the size and composition of
Army A and Army B, it is possible to answer more accurately: Who will win? [1].
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 122–130, 2020.
https://doi.org/10.1007/978-981-15-0187-6_14