Adaptive Game Level Creation through Rank-based Interactive Evolution

Antonios Liapis; Héctor Pérez Martínez; Julian Togelius; Georgios N. Yannakakis

Adaptive Game Level Creation through Rank-based Interactive Evolution

Antonios Liapis, Héctor Pérez Martínez, Julian Togelius, Georgios N. Yannakakis

Center for Digital Play

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

This paper introduces Rank-based Interactive Evolution (RIE) which is an alternative to interactive evolution driven by computational models of user preferences to generate personalized content. In RIE, the computational models are adapted to the preferences of users which, in turn, are used as fitness functions for the optimization of the generated content. The preference models are built via ranking-based preference learning, while the content is generated via evolutionary search. The proposed method is evaluated on the creation of strategy game maps, and its performance is tested using artificial agents. Results suggest that RIE is both faster and more robust than standard interactive evolution and outperforms other state-of-the-art interactive evolution approaches.

Original language	English
Title of host publication	Proceedings of the IEEE Conference on Computational Intelligence and Games (CIG)
Number of pages	8
Publisher	IEEE Computer Society Press
Publication date	2013
Pages	1-8
ISBN (Print)	978-1-4673-5308-3
Publication status	Published - 2013

Keywords

Rank-based Interactive Evolution
User Preference Modeling
Fitness Functions
Evolutionary Search
Strategy Game Maps

Cite this

@inproceedings{3c39f2d8a07943bc843c7cbf2ebe3a2b,

title = "Adaptive Game Level Creation through Rank-based Interactive Evolution",

abstract = "This paper introduces Rank-based Interactive Evolution (RIE) which is an alternative to interactive evolution driven by computational models of user preferences to generate personalized content. In RIE, the computational models are adapted to the preferences of users which, in turn, are used as fitness functions for the optimization of the generated content. The preference models are built via ranking-based preference learning, while the content is generated via evolutionary search. The proposed method is evaluated on the creation of strategy game maps, and its performance is tested using artificial agents. Results suggest that RIE is both faster and more robust than standard interactive evolution and outperforms other state-of-the-art interactive evolution approaches.",

keywords = "Rank-based Interactive Evolution, User Preference Modeling, Fitness Functions, Evolutionary Search, Strategy Game Maps",

author = "Antonios Liapis and Mart{\'i}nez, {H{\'e}ctor P{\'e}rez} and Julian Togelius and Yannakakis, {Georgios N.}",

note = "Der findes dd. kun tentative proceedings: http://geneura.ugr.es/cig2012/proceedings.html 4.1.2014 haal Flyttet til 2014 - ikke udkommet i 2013 22.01.2014 haal",

year = "2013",

language = "English",

isbn = "978-1-4673-5308-3 ",

pages = "1--8",

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

publisher = "IEEE Computer Society Press",

address = "United States",

}

Adaptive Game Level Creation through Rank-based Interactive Evolution. / Liapis, Antonios; Martínez, Héctor Pérez; Togelius, Julian et al.
Proceedings of the IEEE Conference on Computational Intelligence and Games (CIG). IEEE Computer Society Press, 2013. p. 1-8.

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - Adaptive Game Level Creation through Rank-based Interactive Evolution

AU - Liapis, Antonios

AU - Martínez, Héctor Pérez

AU - Togelius, Julian

AU - Yannakakis, Georgios N.

N1 - Der findes dd. kun tentative proceedings: http://geneura.ugr.es/cig2012/proceedings.html 4.1.2014 haal Flyttet til 2014 - ikke udkommet i 2013 22.01.2014 haal

PY - 2013

Y1 - 2013

N2 - This paper introduces Rank-based Interactive Evolution (RIE) which is an alternative to interactive evolution driven by computational models of user preferences to generate personalized content. In RIE, the computational models are adapted to the preferences of users which, in turn, are used as fitness functions for the optimization of the generated content. The preference models are built via ranking-based preference learning, while the content is generated via evolutionary search. The proposed method is evaluated on the creation of strategy game maps, and its performance is tested using artificial agents. Results suggest that RIE is both faster and more robust than standard interactive evolution and outperforms other state-of-the-art interactive evolution approaches.

AB - This paper introduces Rank-based Interactive Evolution (RIE) which is an alternative to interactive evolution driven by computational models of user preferences to generate personalized content. In RIE, the computational models are adapted to the preferences of users which, in turn, are used as fitness functions for the optimization of the generated content. The preference models are built via ranking-based preference learning, while the content is generated via evolutionary search. The proposed method is evaluated on the creation of strategy game maps, and its performance is tested using artificial agents. Results suggest that RIE is both faster and more robust than standard interactive evolution and outperforms other state-of-the-art interactive evolution approaches.

KW - Rank-based Interactive Evolution

KW - User Preference Modeling

KW - Fitness Functions

KW - Evolutionary Search

KW - Strategy Game Maps

M3 - Article in proceedings

SN - 978-1-4673-5308-3

SP - 1

EP - 8

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

PB - IEEE Computer Society Press

ER -

Adaptive Game Level Creation through Rank-based Interactive Evolution

Abstract

Keywords

Fingerprint

Cite this