Game AI generation using evolutionary multi-objective optimization

This paper presents the design and evaluation of a full AI controller for Real-Time Strategy (RTS) games using techniques from Evolutionary Computing (EC). The design is novel in its use of a modified Pareto Differential Evolution (PDE) algorithm for bi-objective optimization of the weights of an Ar...

Full description

Saved in:
Bibliographic Details
Main Authors: Chang Kee Tong, Chin Kim On, Teo, J., Mountstephens, J.
Format: Conference Proceeding
Language:English
Subjects:
Artificial intelligence
Artificial Intelligence (AI)
Artificial neural networks
Artificial Neural Networks (ANN)
Evolutionary Multi-Objective Optimization (EMO)
Games
Humans
Neurons
Optimization
Pareto Differential Evolution (PDE)
Real-Time Strategy Game (RTS)
Vectors
Warcraft III
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 8
container_issue
container_start_page 1
container_title
container_volume
creator Chang Kee Tong
Chin Kim On
Teo, J.
Mountstephens, J.
description This paper presents the design and evaluation of a full AI controller for Real-Time Strategy (RTS) games using techniques from Evolutionary Computing (EC). The design is novel in its use of a modified Pareto Differential Evolution (PDE) algorithm for bi-objective optimization of the weights of an Artificial Neural Network (ANN) controller when only single-objective optimization has so far been studied. The two main aims of this research are to: (1) develop controllers capable of defeating opponents of varying difficulty levels, which may assist in commercial RTS AI development, and (2) minimize the number of neurons used in the ANN architecture, an issue primarily of efficiency. Experimental results using the popular Warcraft III platform demonstrate success with both aims: the optimized controller was able to win any battle using only a minimal number of hidden neurons, but sub-optimal controllers were able to provide opponents of any intermediate difficulty.
doi_str_mv 10.1109/CEC.2012.6256638
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_6256638</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6256638</ieee_id><sourcerecordid>6256638</sourcerecordid><originalsourceid>FETCH-LOGICAL-i175t-58cee2d7c3d421ea09f0181d8d0976e71d1385353eee76f4fb2831037f778eec3</originalsourceid><addsrcrecordid>eNo9kElLA0EQhdsNjDF3wUv_gR6ruqeXOcYhxkDAi4K3MJmpCR1mCbME9Nen1WBdHsX7qng8xh4QIkRIntJFGklAGRmpjVHugt1hbKxCDc5esgkmMQoAaa7-jXB2HQxwibDWfd6yWd_vIYx1iLGdsOdlVhOfr_iOGuqywbcNH3vf7Dgd22r82bPui9djNXjRbveUD_5IvD0Mvvbfv_w9uymzqqfZWafs42Xxnr6K9dtylc7XwqPVg9AuJ5KFzVURS6QMkhLQYeEKSKwhiwUqp5VWRGRNGZdb6RSCsmVITpSrKXv8--sDsTl0vg7JNucu1AkD1E6I</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Game AI generation using evolutionary multi-objective optimization</title><source>IEEE Xplore All Conference Series</source><creator>Chang Kee Tong ; Chin Kim On ; Teo, J. ; Mountstephens, J.</creator><creatorcontrib>Chang Kee Tong ; Chin Kim On ; Teo, J. ; Mountstephens, J.</creatorcontrib><description>This paper presents the design and evaluation of a full AI controller for Real-Time Strategy (RTS) games using techniques from Evolutionary Computing (EC). The design is novel in its use of a modified Pareto Differential Evolution (PDE) algorithm for bi-objective optimization of the weights of an Artificial Neural Network (ANN) controller when only single-objective optimization has so far been studied. The two main aims of this research are to: (1) develop controllers capable of defeating opponents of varying difficulty levels, which may assist in commercial RTS AI development, and (2) minimize the number of neurons used in the ANN architecture, an issue primarily of efficiency. Experimental results using the popular Warcraft III platform demonstrate success with both aims: the optimized controller was able to win any battle using only a minimal number of hidden neurons, but sub-optimal controllers were able to provide opponents of any intermediate difficulty.</description><identifier>ISSN: 1089-778X</identifier><identifier>ISBN: 1467315109</identifier><identifier>ISBN: 9781467315104</identifier><identifier>EISSN: 1941-0026</identifier><identifier>EISBN: 1467315087</identifier><identifier>EISBN: 9781467315081</identifier><identifier>EISBN: 1467315095</identifier><identifier>EISBN: 9781467315098</identifier><identifier>DOI: 10.1109/CEC.2012.6256638</identifier><language>eng</language><publisher>IEEE</publisher><subject>Artificial intelligence ; Artificial Intelligence (AI) ; Artificial neural networks ; Artificial Neural Networks (ANN) ; Evolutionary Multi-Objective Optimization (EMO) ; Games ; Humans ; Neurons ; Optimization ; Pareto Differential Evolution (PDE) ; Real-Time Strategy Game (RTS) ; Vectors ; Warcraft III</subject><ispartof>2012 IEEE Congress on Evolutionary Computation, 2012, p.1-8</ispartof><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6256638$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2058,27925,54555,54796,54920,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6256638$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Chang Kee Tong</creatorcontrib><creatorcontrib>Chin Kim On</creatorcontrib><creatorcontrib>Teo, J.</creatorcontrib><creatorcontrib>Mountstephens, J.</creatorcontrib><title>Game AI generation using evolutionary multi-objective optimization</title><title>2012 IEEE Congress on Evolutionary Computation</title><addtitle>CEC</addtitle><description>This paper presents the design and evaluation of a full AI controller for Real-Time Strategy (RTS) games using techniques from Evolutionary Computing (EC). The design is novel in its use of a modified Pareto Differential Evolution (PDE) algorithm for bi-objective optimization of the weights of an Artificial Neural Network (ANN) controller when only single-objective optimization has so far been studied. The two main aims of this research are to: (1) develop controllers capable of defeating opponents of varying difficulty levels, which may assist in commercial RTS AI development, and (2) minimize the number of neurons used in the ANN architecture, an issue primarily of efficiency. Experimental results using the popular Warcraft III platform demonstrate success with both aims: the optimized controller was able to win any battle using only a minimal number of hidden neurons, but sub-optimal controllers were able to provide opponents of any intermediate difficulty.</description><subject>Artificial intelligence</subject><subject>Artificial Intelligence (AI)</subject><subject>Artificial neural networks</subject><subject>Artificial Neural Networks (ANN)</subject><subject>Evolutionary Multi-Objective Optimization (EMO)</subject><subject>Games</subject><subject>Humans</subject><subject>Neurons</subject><subject>Optimization</subject><subject>Pareto Differential Evolution (PDE)</subject><subject>Real-Time Strategy Game (RTS)</subject><subject>Vectors</subject><subject>Warcraft III</subject><issn>1089-778X</issn><issn>1941-0026</issn><isbn>1467315109</isbn><isbn>9781467315104</isbn><isbn>1467315087</isbn><isbn>9781467315081</isbn><isbn>1467315095</isbn><isbn>9781467315098</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2012</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNo9kElLA0EQhdsNjDF3wUv_gR6ruqeXOcYhxkDAi4K3MJmpCR1mCbME9Nen1WBdHsX7qng8xh4QIkRIntJFGklAGRmpjVHugt1hbKxCDc5esgkmMQoAaa7-jXB2HQxwibDWfd6yWd_vIYx1iLGdsOdlVhOfr_iOGuqywbcNH3vf7Dgd22r82bPui9djNXjRbveUD_5IvD0Mvvbfv_w9uymzqqfZWafs42Xxnr6K9dtylc7XwqPVg9AuJ5KFzVURS6QMkhLQYeEKSKwhiwUqp5VWRGRNGZdb6RSCsmVITpSrKXv8--sDsTl0vg7JNucu1AkD1E6I</recordid><startdate>201206</startdate><enddate>201206</enddate><creator>Chang Kee Tong</creator><creator>Chin Kim On</creator><creator>Teo, J.</creator><creator>Mountstephens, J.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>201206</creationdate><title>Game AI generation using evolutionary multi-objective optimization</title><author>Chang Kee Tong ; Chin Kim On ; Teo, J. ; Mountstephens, J.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i175t-58cee2d7c3d421ea09f0181d8d0976e71d1385353eee76f4fb2831037f778eec3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Artificial intelligence</topic><topic>Artificial Intelligence (AI)</topic><topic>Artificial neural networks</topic><topic>Artificial Neural Networks (ANN)</topic><topic>Evolutionary Multi-Objective Optimization (EMO)</topic><topic>Games</topic><topic>Humans</topic><topic>Neurons</topic><topic>Optimization</topic><topic>Pareto Differential Evolution (PDE)</topic><topic>Real-Time Strategy Game (RTS)</topic><topic>Vectors</topic><topic>Warcraft III</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Chang Kee Tong</creatorcontrib><creatorcontrib>Chin Kim On</creatorcontrib><creatorcontrib>Teo, J.</creatorcontrib><creatorcontrib>Mountstephens, J.</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE/IET Electronic Library</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Chang Kee Tong</au><au>Chin Kim On</au><au>Teo, J.</au><au>Mountstephens, J.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Game AI generation using evolutionary multi-objective optimization</atitle><btitle>2012 IEEE Congress on Evolutionary Computation</btitle><stitle>CEC</stitle><date>2012-06</date><risdate>2012</risdate><spage>1</spage><epage>8</epage><pages>1-8</pages><issn>1089-778X</issn><eissn>1941-0026</eissn><isbn>1467315109</isbn><isbn>9781467315104</isbn><eisbn>1467315087</eisbn><eisbn>9781467315081</eisbn><eisbn>1467315095</eisbn><eisbn>9781467315098</eisbn><abstract>This paper presents the design and evaluation of a full AI controller for Real-Time Strategy (RTS) games using techniques from Evolutionary Computing (EC). The design is novel in its use of a modified Pareto Differential Evolution (PDE) algorithm for bi-objective optimization of the weights of an Artificial Neural Network (ANN) controller when only single-objective optimization has so far been studied. The two main aims of this research are to: (1) develop controllers capable of defeating opponents of varying difficulty levels, which may assist in commercial RTS AI development, and (2) minimize the number of neurons used in the ANN architecture, an issue primarily of efficiency. Experimental results using the popular Warcraft III platform demonstrate success with both aims: the optimized controller was able to win any battle using only a minimal number of hidden neurons, but sub-optimal controllers were able to provide opponents of any intermediate difficulty.</abstract><pub>IEEE</pub><doi>10.1109/CEC.2012.6256638</doi><tpages>8</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 1089-778X
ispartof 2012 IEEE Congress on Evolutionary Computation, 2012, p.1-8
issn 1089-778X
1941-0026
language eng
recordid cdi_ieee_primary_6256638
source IEEE Xplore All Conference Series
subjects Artificial intelligence
Artificial Intelligence (AI)
Artificial neural networks
Artificial Neural Networks (ANN)
Evolutionary Multi-Objective Optimization (EMO)
Games
Humans
Neurons
Optimization
Pareto Differential Evolution (PDE)
Real-Time Strategy Game (RTS)
Vectors
Warcraft III
title Game AI generation using evolutionary multi-objective optimization
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-29T20%3A29%3A26IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Game%20AI%20generation%20using%20evolutionary%20multi-objective%20optimization&rft.btitle=2012%20IEEE%20Congress%20on%20Evolutionary%20Computation&rft.au=Chang%20Kee%20Tong&rft.date=2012-06&rft.spage=1&rft.epage=8&rft.pages=1-8&rft.issn=1089-778X&rft.eissn=1941-0026&rft.isbn=1467315109&rft.isbn_list=9781467315104&rft_id=info:doi/10.1109/CEC.2012.6256638&rft.eisbn=1467315087&rft.eisbn_list=9781467315081&rft.eisbn_list=1467315095&rft.eisbn_list=9781467315098&rft_dat=%3Cieee_CHZPO%3E6256638%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i175t-58cee2d7c3d421ea09f0181d8d0976e71d1385353eee76f4fb2831037f778eec3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=6256638&rfr_iscdi=true