Deep learning for procedural content generation

Jialin LIU; Sam SNODGRASS; Ahmed KHALIFA; Sebastian RISI; Georgios N. YANNAKAKIS; Julian TOGELIUS

doi:10.1007/s00521-020-05383-8

Deep learning for procedural content generation

Jialin LIU, Sam SNODGRASS, Ahmed KHALIFA, Sebastian RISI, Georgios N. YANNAKAKIS, Julian TOGELIUS^*

^*Corresponding author for this work

Research output: Journal Publications › Journal Article (refereed) › peer-review

64 Citations (Scopus)

Abstract

Procedural content generation in video games has a long history. Existing procedural content generation methods, such as search-based, solver-based, rule-based and grammar-based methods have been applied to various content types such as levels, maps, character models, and textures. A research field centered on content generation in games has existed for more than a decade. More recently, deep learning has powered a remarkable range of inventions in content production, which are applicable to games. While some cutting-edge deep learning methods are applied on their own, others are applied in combination with more traditional methods, or in an interactive setting. This article surveys the various deep learning methods that have been applied to generate game content directly or indirectly, discusses deep learning methods that could be used for content generation purposes but are rarely used today, and envisages some limitations and potential future directions of deep learning for procedural content generation.

Original language	English
Pages (from-to)	19-37
Number of pages	19
Journal	Neural Computing and Applications
Volume	33
Issue number	1
Early online date	8 Oct 2020
DOIs	https://doi.org/10.1007/s00521-020-05383-8
Publication status	Published - Jan 2021
Externally published	Yes

Bibliographical note

J. Liu was supported by the National Key R&D Program of China (Grant No. 2017YFC0804003), the National Natural Science Foundation of China (Grant No. 61906083), the Guangdong Provincial Key Laboratory (Grant No. 2020B121201001), the Program for Guangdong Introducing Innovative and Entrepreneurial Teams (Grant No. 2017ZT07X386), the Science and Technology Innovation Committee Foundation of Shenzhen (Grant No. JCYJ20190809121403553), the Shenzhen Science and Technology Program (Grant No. KQTD2016112514355531) and the Program for University Key Laboratory of Guangdong Province (Grant No. 2017KSYS008). S. Risi was supported by a Google Faculty Research award and a Sapere Aude: DFF-Starting Grant. A. Khalifa and J. Togelius acknowledge the financial support from National Science Foundation (NSF) award number 1717324 - “RI: Small: General Intelligence through Algorithm Invention and Selection”. G. N. Yannakakis was supported by European Union’s Horizon 2020 AI4Media (951911) and TAMED (101003397) projects.

Keywords

Computational and artificial intelligence
Deep learning
Game design
Machine learning
Procedural content generation

Access to Document

10.1007/s00521-020-05383-8

Cite this

@article{1e31101ddb334d408f19d662f18bd529,

title = "Deep learning for procedural content generation",

abstract = "Procedural content generation in video games has a long history. Existing procedural content generation methods, such as search-based, solver-based, rule-based and grammar-based methods have been applied to various content types such as levels, maps, character models, and textures. A research field centered on content generation in games has existed for more than a decade. More recently, deep learning has powered a remarkable range of inventions in content production, which are applicable to games. While some cutting-edge deep learning methods are applied on their own, others are applied in combination with more traditional methods, or in an interactive setting. This article surveys the various deep learning methods that have been applied to generate game content directly or indirectly, discusses deep learning methods that could be used for content generation purposes but are rarely used today, and envisages some limitations and potential future directions of deep learning for procedural content generation.",

keywords = "Computational and artificial intelligence, Deep learning, Game design, Machine learning, Procedural content generation",

author = "Jialin LIU and Sam SNODGRASS and Ahmed KHALIFA and Sebastian RISI and YANNAKAKIS, {Georgios N.} and Julian TOGELIUS",

note = "J. Liu was supported by the National Key R&D Program of China (Grant No. 2017YFC0804003), the National Natural Science Foundation of China (Grant No. 61906083), the Guangdong Provincial Key Laboratory (Grant No. 2020B121201001), the Program for Guangdong Introducing Innovative and Entrepreneurial Teams (Grant No. 2017ZT07X386), the Science and Technology Innovation Committee Foundation of Shenzhen (Grant No. JCYJ20190809121403553), the Shenzhen Science and Technology Program (Grant No. KQTD2016112514355531) and the Program for University Key Laboratory of Guangdong Province (Grant No. 2017KSYS008). S. Risi was supported by a Google Faculty Research award and a Sapere Aude: DFF-Starting Grant. A. Khalifa and J. Togelius acknowledge the financial support from National Science Foundation (NSF) award number 1717324 - “RI: Small: General Intelligence through Algorithm Invention and Selection”. G. N. Yannakakis was supported by European Union{\textquoteright}s Horizon 2020 AI4Media (951911) and TAMED (101003397) projects.",

year = "2021",

month = jan,

doi = "10.1007/s00521-020-05383-8",

language = "English",

volume = "33",

pages = "19--37",

journal = "Neural Computing and Applications",

issn = "0941-0643",

publisher = "Springer London",

number = "1",

}

TY - JOUR

T1 - Deep learning for procedural content generation

AU - LIU, Jialin

AU - SNODGRASS, Sam

AU - KHALIFA, Ahmed

AU - RISI, Sebastian

AU - YANNAKAKIS, Georgios N.

AU - TOGELIUS, Julian

N1 - J. Liu was supported by the National Key R&D Program of China (Grant No. 2017YFC0804003), the National Natural Science Foundation of China (Grant No. 61906083), the Guangdong Provincial Key Laboratory (Grant No. 2020B121201001), the Program for Guangdong Introducing Innovative and Entrepreneurial Teams (Grant No. 2017ZT07X386), the Science and Technology Innovation Committee Foundation of Shenzhen (Grant No. JCYJ20190809121403553), the Shenzhen Science and Technology Program (Grant No. KQTD2016112514355531) and the Program for University Key Laboratory of Guangdong Province (Grant No. 2017KSYS008). S. Risi was supported by a Google Faculty Research award and a Sapere Aude: DFF-Starting Grant. A. Khalifa and J. Togelius acknowledge the financial support from National Science Foundation (NSF) award number 1717324 - “RI: Small: General Intelligence through Algorithm Invention and Selection”. G. N. Yannakakis was supported by European Union’s Horizon 2020 AI4Media (951911) and TAMED (101003397) projects.

PY - 2021/1

Y1 - 2021/1

N2 - Procedural content generation in video games has a long history. Existing procedural content generation methods, such as search-based, solver-based, rule-based and grammar-based methods have been applied to various content types such as levels, maps, character models, and textures. A research field centered on content generation in games has existed for more than a decade. More recently, deep learning has powered a remarkable range of inventions in content production, which are applicable to games. While some cutting-edge deep learning methods are applied on their own, others are applied in combination with more traditional methods, or in an interactive setting. This article surveys the various deep learning methods that have been applied to generate game content directly or indirectly, discusses deep learning methods that could be used for content generation purposes but are rarely used today, and envisages some limitations and potential future directions of deep learning for procedural content generation.

AB - Procedural content generation in video games has a long history. Existing procedural content generation methods, such as search-based, solver-based, rule-based and grammar-based methods have been applied to various content types such as levels, maps, character models, and textures. A research field centered on content generation in games has existed for more than a decade. More recently, deep learning has powered a remarkable range of inventions in content production, which are applicable to games. While some cutting-edge deep learning methods are applied on their own, others are applied in combination with more traditional methods, or in an interactive setting. This article surveys the various deep learning methods that have been applied to generate game content directly or indirectly, discusses deep learning methods that could be used for content generation purposes but are rarely used today, and envisages some limitations and potential future directions of deep learning for procedural content generation.

KW - Computational and artificial intelligence

KW - Deep learning

KW - Game design

KW - Machine learning

KW - Procedural content generation

UR - http://www.scopus.com/inward/record.url?scp=85092349783&partnerID=8YFLogxK

U2 - 10.1007/s00521-020-05383-8

DO - 10.1007/s00521-020-05383-8

M3 - Journal Article (refereed)

AN - SCOPUS:85092349783

SN - 0941-0643

VL - 33

SP - 19

EP - 37

JO - Neural Computing and Applications

JF - Neural Computing and Applications

IS - 1

ER -

Deep learning for procedural content generation

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this