are common used clustering algorithms [31]. Besides, popular frequent pattern mining
methods include the Apriori algorithm for itemset mining, and SPADE [32] and GSP
[33] for sequence mining.
6 Conclusions
As the list in the previous section shows that games are an ideal domain for AI, which
poses a large number of open problems. In this paper, we review the development of
game AI. Different AI methods and their applications in games are presented. Obviously, it is very difficult to design a universal game AI with a unified framework.
Different tasks represent different abstractions and involve different time spans.
Therefore, a highly intelligent game AI needs the combination of various AI methods
in order to learn from each other’s strong points, and the emergence of notable hybrid
algorithms such as neuroevolution and deep Q-learning is strong evidence.
References
1. Cruz RMO, Sabourin R, Cavalcanti GDC et al (2015) META-DES: a dynamic ensemble
selection framework using meta-learning. Pattern Recognit 48(5):1925–1935
2. Schaeffer J, Lake R, Lu P et al (1996) Chinook: the world man machine checkers champion.
AI Mag 17(1):21
3. Campbell M, Hoane AJ, Hsu F (2002) Deep blue. Artif Intell 134(1):57–83
4. Silver D, Huang A, Maddison CJ et al (2016) Mastering the game of Go with deep neural
networks and tree search. Nature 529(7587):484–489
5. Mnih V, Kavukcuoglu K, Silver D et al (2015) Human-level control through deep
reinforcement learning. Nature 518(7540):529–533
6. Tang Z, Shao K, Zhu Y et al (2018) A review of computational intelligence for StarCraft AI.
In: IEEE SSCI, pp 1167–1173
7. Ontañón S, Synnaeve G, Uriarte A et al (2013) A survey of real-time strategy game AI
research and competition in StarCraft. IEEE Trans Comput Intell AI 5(4):293–311
8. Ketter W, Collins J, Gini M et al (2012) Real-time tactical and strategic sales management
for intelligent agents guided by economic regimes. Inf Syst Res 23(4):1263–1283
9. Hidalgo-Bermúdez RM, Rodríguez-Domingo MS, Mora AM et al (2013) Evolutionary
FSM-based agents for playing super Mario game. LION. https://doi.org/10.1007/978-3-64244973-4_39
10. Toll WGV, Iv AFC, Geraerts R et al (2012) Real-time density-based crowd simulation.
Comput Animat Virtual Worlds 23(1):59–69
11. Jadon S, Singhal A, Dawn S (2014) Military simulator—a case study of behaviour tree and
unity based architecture. Int J Comput Appl 88(5):26–29
12. Guillaume MJ, Mark HM, Herik HJ et al (2008) Progressive strategies for Monte-Carlo tree
search. NMNC 4(03):343–357
13. Coquelin R, Munos A (2007) Bandit algorithms for tree search. In: Conference on
uncertainty in artificial intelligence, pp 67–74
14. Silver D, Sutton RS, Müller M (2008) Sample-based learning and search with permanent and
transient memories. In: Annual international conference on machine learning, pp 968–975
302
X. Fan et al.
methods include the Apriori algorithm for itemset mining, and SPADE [32] and GSP
[33] for sequence mining.
6 Conclusions
As the list in the previous section shows that games are an ideal domain for AI, which
poses a large number of open problems. In this paper, we review the development of
game AI. Different AI methods and their applications in games are presented. Obviously, it is very difficult to design a universal game AI with a unified framework.
Different tasks represent different abstractions and involve different time spans.
Therefore, a highly intelligent game AI needs the combination of various AI methods
in order to learn from each other’s strong points, and the emergence of notable hybrid
algorithms such as neuroevolution and deep Q-learning is strong evidence.
References
1. Cruz RMO, Sabourin R, Cavalcanti GDC et al (2015) META-DES: a dynamic ensemble
selection framework using meta-learning. Pattern Recognit 48(5):1925–1935
2. Schaeffer J, Lake R, Lu P et al (1996) Chinook: the world man machine checkers champion.
AI Mag 17(1):21
3. Campbell M, Hoane AJ, Hsu F (2002) Deep blue. Artif Intell 134(1):57–83
4. Silver D, Huang A, Maddison CJ et al (2016) Mastering the game of Go with deep neural
networks and tree search. Nature 529(7587):484–489
5. Mnih V, Kavukcuoglu K, Silver D et al (2015) Human-level control through deep
reinforcement learning. Nature 518(7540):529–533
6. Tang Z, Shao K, Zhu Y et al (2018) A review of computational intelligence for StarCraft AI.
In: IEEE SSCI, pp 1167–1173
7. Ontañón S, Synnaeve G, Uriarte A et al (2013) A survey of real-time strategy game AI
research and competition in StarCraft. IEEE Trans Comput Intell AI 5(4):293–311
8. Ketter W, Collins J, Gini M et al (2012) Real-time tactical and strategic sales management
for intelligent agents guided by economic regimes. Inf Syst Res 23(4):1263–1283
9. Hidalgo-Bermúdez RM, Rodríguez-Domingo MS, Mora AM et al (2013) Evolutionary
FSM-based agents for playing super Mario game. LION. https://doi.org/10.1007/978-3-64244973-4_39
10. Toll WGV, Iv AFC, Geraerts R et al (2012) Real-time density-based crowd simulation.
Comput Animat Virtual Worlds 23(1):59–69
11. Jadon S, Singhal A, Dawn S (2014) Military simulator—a case study of behaviour tree and
unity based architecture. Int J Comput Appl 88(5):26–29
12. Guillaume MJ, Mark HM, Herik HJ et al (2008) Progressive strategies for Monte-Carlo tree
search. NMNC 4(03):343–357
13. Coquelin R, Munos A (2007) Bandit algorithms for tree search. In: Conference on
uncertainty in artificial intelligence, pp 67–74
14. Silver D, Sutton RS, Müller M (2008) Sample-based learning and search with permanent and
transient memories. In: Annual international conference on machine learning, pp 968–975
302
X. Fan et al.