Towards Adaptive Online RTS AI with NEAT

Jason Traish; James Tulip

doi:10.1109/CIG.2012.6374187

Towards Adaptive Online RTS AI with NEAT

Jason Traish, James Tulip

Data Science Research Unit

Research output: Book chapter/Published conference paper › Conference paper › peer-review

6 Citations (Scopus)

101 Downloads (Pure)

Abstract

Real Time Strategy (RTS) games are interesting from an Artificial Intelligence (AI) point of view because they involve a huge range of decision making from local tactical decisions to broad strategic considerations, all of which occur on a densely populated and fiercely contested map. However, most RTS AI used in commercial RTS games are predictable and can be exploited by expert players.Adaptive or evolutionary AI techniques offer the potential to create challenging AI opponents. Neural Evolution of Augmenting Technologies (NEAT) is a hybrid approach that applies Genetic Algorithm (GA) techniques to increase the efficiency of learning neural nets. This work presents an application of NEAT to RTS AI. It does so through a set of experiments in a realistic RTS environment.The results of the experiments show that NEAT can produce satisfactory RTS agents, and can also create agents capable of displaying complex in-game adaptive behavior. The results are significant because they show that NEAT can be used to evolve sophisticated RTS AI opponents without significant designer input or expertise, and without extensive databases of existing games.

Original language	English
Title of host publication	Proceedings of the 2012 IEEE Conference on Computational Intelligence and Games (CIG)
Place of Publication	United States
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	430-437
Number of pages	8
ISBN (Electronic)	9781467311939
DOIs	https://doi.org/10.1109/CIG.2012.6374187
Publication status	Published - 2012
Event	2012 IEEE International Conference on Computational Intelligence and Games: CIG 2012 - Aynadamar Campus, University of Granada, Granada, Spain Duration: 11 Sept 2012 → 14 Sept 2012

Conference

Conference	2012 IEEE International Conference on Computational Intelligence and Games
Country/Territory	Spain
City	Granada
Period	11/09/12 → 14/09/12

Access to Document

10.1109/CIG.2012.6374187

41227 ManuscriptAccepted author manuscript, 679 KB

Cite this

@inproceedings{32090a5d43c04217b7572ec016427551,

title = "Towards Adaptive Online RTS AI with NEAT",

abstract = "Real Time Strategy (RTS) games are interesting from an Artificial Intelligence (AI) point of view because they involve a huge range of decision making from local tactical decisions to broad strategic considerations, all of which occur on a densely populated and fiercely contested map. However, most RTS AI used in commercial RTS games are predictable and can be exploited by expert players.Adaptive or evolutionary AI techniques offer the potential to create challenging AI opponents. Neural Evolution of Augmenting Technologies (NEAT) is a hybrid approach that applies Genetic Algorithm (GA) techniques to increase the efficiency of learning neural nets. This work presents an application of NEAT to RTS AI. It does so through a set of experiments in a realistic RTS environment.The results of the experiments show that NEAT can produce satisfactory RTS agents, and can also create agents capable of displaying complex in-game adaptive behavior. The results are significant because they show that NEAT can be used to evolve sophisticated RTS AI opponents without significant designer input or expertise, and without extensive databases of existing games.",

keywords = "Open access version available, AI, Adaptive, Evolutionary, NEAT, RTS, Realtime strategy game",

author = "Jason Traish and James Tulip",

note = "Imported on 03 May 2017 - DigiTool details were: publisher = United States: Institute of Electrical and Electronics Engineers, 2012. editor/s (773b) = CIG 2012 proceedings; Event dates (773o) = 11-14 September 2012; Parent title (773t) = IEEE Symposium on Computational Intelligence and Games.; 2012 IEEE International Conference on Computational Intelligence and Games : CIG 2012 ; Conference date: 11-09-2012 Through 14-09-2012",

year = "2012",

doi = "10.1109/CIG.2012.6374187",

language = "English",

pages = "430--437",

booktitle = "Proceedings of the 2012 IEEE Conference on Computational Intelligence and Games (CIG)",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States",

}

Traish, J & Tulip, J 2012, Towards Adaptive Online RTS AI with NEAT. in Proceedings of the 2012 IEEE Conference on Computational Intelligence and Games (CIG). IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 430-437, 2012 IEEE International Conference on Computational Intelligence and Games, Granada, Spain, 11/09/12. https://doi.org/10.1109/CIG.2012.6374187

TY - GEN

T1 - Towards Adaptive Online RTS AI with NEAT

AU - Traish, Jason

AU - Tulip, James

N1 - Imported on 03 May 2017 - DigiTool details were: publisher = United States: Institute of Electrical and Electronics Engineers, 2012. editor/s (773b) = CIG 2012 proceedings; Event dates (773o) = 11-14 September 2012; Parent title (773t) = IEEE Symposium on Computational Intelligence and Games.

PY - 2012

Y1 - 2012

N2 - Real Time Strategy (RTS) games are interesting from an Artificial Intelligence (AI) point of view because they involve a huge range of decision making from local tactical decisions to broad strategic considerations, all of which occur on a densely populated and fiercely contested map. However, most RTS AI used in commercial RTS games are predictable and can be exploited by expert players.Adaptive or evolutionary AI techniques offer the potential to create challenging AI opponents. Neural Evolution of Augmenting Technologies (NEAT) is a hybrid approach that applies Genetic Algorithm (GA) techniques to increase the efficiency of learning neural nets. This work presents an application of NEAT to RTS AI. It does so through a set of experiments in a realistic RTS environment.The results of the experiments show that NEAT can produce satisfactory RTS agents, and can also create agents capable of displaying complex in-game adaptive behavior. The results are significant because they show that NEAT can be used to evolve sophisticated RTS AI opponents without significant designer input or expertise, and without extensive databases of existing games.

AB - Real Time Strategy (RTS) games are interesting from an Artificial Intelligence (AI) point of view because they involve a huge range of decision making from local tactical decisions to broad strategic considerations, all of which occur on a densely populated and fiercely contested map. However, most RTS AI used in commercial RTS games are predictable and can be exploited by expert players.Adaptive or evolutionary AI techniques offer the potential to create challenging AI opponents. Neural Evolution of Augmenting Technologies (NEAT) is a hybrid approach that applies Genetic Algorithm (GA) techniques to increase the efficiency of learning neural nets. This work presents an application of NEAT to RTS AI. It does so through a set of experiments in a realistic RTS environment.The results of the experiments show that NEAT can produce satisfactory RTS agents, and can also create agents capable of displaying complex in-game adaptive behavior. The results are significant because they show that NEAT can be used to evolve sophisticated RTS AI opponents without significant designer input or expertise, and without extensive databases of existing games.

KW - Open access version available

KW - AI

KW - Adaptive

KW - Evolutionary

KW - NEAT

KW - RTS

KW - Realtime strategy game

U2 - 10.1109/CIG.2012.6374187

DO - 10.1109/CIG.2012.6374187

M3 - Conference paper

SP - 430

EP - 437

BT - Proceedings of the 2012 IEEE Conference on Computational Intelligence and Games (CIG)

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

T2 - 2012 IEEE International Conference on Computational Intelligence and Games

Y2 - 11 September 2012 through 14 September 2012

ER -

Towards Adaptive Online RTS AI with NEAT

Abstract

Conference

Access to Document

Fingerprint

Cite this