Learning Models of Adversarial Agent Behavior under Partial Observability
The need for opponent modeling and tracking arises in several real-world scenarios, such as professional sports, video game design, and drug-trafficking interdiction. In this work, we present Graph based Adversarial Modeling with Mutal Information (GrAMMI) for modeling the behavior of an adversarial...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2023-07 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Ye, Sean Natarajan, Manisha Wu, Zixuan Paleja, Rohan Chen, Letian Gombolay, Matthew C |
description | The need for opponent modeling and tracking arises in several real-world scenarios, such as professional sports, video game design, and drug-trafficking interdiction. In this work, we present Graph based Adversarial Modeling with Mutal Information (GrAMMI) for modeling the behavior of an adversarial opponent agent. GrAMMI is a novel graph neural network (GNN) based approach that uses mutual information maximization as an auxiliary objective to predict the current and future states of an adversarial opponent with partial observability. To evaluate GrAMMI, we design two large-scale, pursuit-evasion domains inspired by real-world scenarios, where a team of heterogeneous agents is tasked with tracking and interdicting a single adversarial agent, and the adversarial agent must evade detection while achieving its own objectives. With the mutual information formulation, GrAMMI outperforms all baselines in both domains and achieves 31.68% higher log-likelihood on average for future adversarial state predictions across both domains. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2828072210</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2828072210</sourcerecordid><originalsourceid>FETCH-proquest_journals_28280722103</originalsourceid><addsrcrecordid>eNqNytEKgjAUgOERBEn5DoOuhXmW6a1FUVDURfcy8WiTsdXZFHr7CnqArv6L75-wCKRMk2IFMGOx970QAtY5ZJmM2PGEiqy2HT-7Bo3nruVlMyJ5RVoZXnZoA9_gXY3aER9sg8SvisIXL7VHGlWtjQ6vBZu2yniMf52z5X532x6SB7nngD5UvRvIfqiCAgqRA6RC_ne9AXNuPE4</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2828072210</pqid></control><display><type>article</type><title>Learning Models of Adversarial Agent Behavior under Partial Observability</title><source>Free E- Journals</source><creator>Ye, Sean ; Natarajan, Manisha ; Wu, Zixuan ; Paleja, Rohan ; Chen, Letian ; Gombolay, Matthew C</creator><creatorcontrib>Ye, Sean ; Natarajan, Manisha ; Wu, Zixuan ; Paleja, Rohan ; Chen, Letian ; Gombolay, Matthew C</creatorcontrib><description>The need for opponent modeling and tracking arises in several real-world scenarios, such as professional sports, video game design, and drug-trafficking interdiction. In this work, we present Graph based Adversarial Modeling with Mutal Information (GrAMMI) for modeling the behavior of an adversarial opponent agent. GrAMMI is a novel graph neural network (GNN) based approach that uses mutual information maximization as an auxiliary objective to predict the current and future states of an adversarial opponent with partial observability. To evaluate GrAMMI, we design two large-scale, pursuit-evasion domains inspired by real-world scenarios, where a team of heterogeneous agents is tasked with tracking and interdicting a single adversarial agent, and the adversarial agent must evade detection while achieving its own objectives. With the mutual information formulation, GrAMMI outperforms all baselines in both domains and achieves 31.68% higher log-likelihood on average for future adversarial state predictions across both domains.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Computer & video games ; Domains ; Graph neural networks ; Modelling ; Pursuit-evasion games ; Tracking</subject><ispartof>arXiv.org, 2023-07</ispartof><rights>2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Ye, Sean</creatorcontrib><creatorcontrib>Natarajan, Manisha</creatorcontrib><creatorcontrib>Wu, Zixuan</creatorcontrib><creatorcontrib>Paleja, Rohan</creatorcontrib><creatorcontrib>Chen, Letian</creatorcontrib><creatorcontrib>Gombolay, Matthew C</creatorcontrib><title>Learning Models of Adversarial Agent Behavior under Partial Observability</title><title>arXiv.org</title><description>The need for opponent modeling and tracking arises in several real-world scenarios, such as professional sports, video game design, and drug-trafficking interdiction. In this work, we present Graph based Adversarial Modeling with Mutal Information (GrAMMI) for modeling the behavior of an adversarial opponent agent. GrAMMI is a novel graph neural network (GNN) based approach that uses mutual information maximization as an auxiliary objective to predict the current and future states of an adversarial opponent with partial observability. To evaluate GrAMMI, we design two large-scale, pursuit-evasion domains inspired by real-world scenarios, where a team of heterogeneous agents is tasked with tracking and interdicting a single adversarial agent, and the adversarial agent must evade detection while achieving its own objectives. With the mutual information formulation, GrAMMI outperforms all baselines in both domains and achieves 31.68% higher log-likelihood on average for future adversarial state predictions across both domains.</description><subject>Computer & video games</subject><subject>Domains</subject><subject>Graph neural networks</subject><subject>Modelling</subject><subject>Pursuit-evasion games</subject><subject>Tracking</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNytEKgjAUgOERBEn5DoOuhXmW6a1FUVDURfcy8WiTsdXZFHr7CnqArv6L75-wCKRMk2IFMGOx970QAtY5ZJmM2PGEiqy2HT-7Bo3nruVlMyJ5RVoZXnZoA9_gXY3aER9sg8SvisIXL7VHGlWtjQ6vBZu2yniMf52z5X532x6SB7nngD5UvRvIfqiCAgqRA6RC_ne9AXNuPE4</recordid><startdate>20230705</startdate><enddate>20230705</enddate><creator>Ye, Sean</creator><creator>Natarajan, Manisha</creator><creator>Wu, Zixuan</creator><creator>Paleja, Rohan</creator><creator>Chen, Letian</creator><creator>Gombolay, Matthew C</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20230705</creationdate><title>Learning Models of Adversarial Agent Behavior under Partial Observability</title><author>Ye, Sean ; Natarajan, Manisha ; Wu, Zixuan ; Paleja, Rohan ; Chen, Letian ; Gombolay, Matthew C</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28280722103</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer & video games</topic><topic>Domains</topic><topic>Graph neural networks</topic><topic>Modelling</topic><topic>Pursuit-evasion games</topic><topic>Tracking</topic><toplevel>online_resources</toplevel><creatorcontrib>Ye, Sean</creatorcontrib><creatorcontrib>Natarajan, Manisha</creatorcontrib><creatorcontrib>Wu, Zixuan</creatorcontrib><creatorcontrib>Paleja, Rohan</creatorcontrib><creatorcontrib>Chen, Letian</creatorcontrib><creatorcontrib>Gombolay, Matthew C</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ye, Sean</au><au>Natarajan, Manisha</au><au>Wu, Zixuan</au><au>Paleja, Rohan</au><au>Chen, Letian</au><au>Gombolay, Matthew C</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Learning Models of Adversarial Agent Behavior under Partial Observability</atitle><jtitle>arXiv.org</jtitle><date>2023-07-05</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>The need for opponent modeling and tracking arises in several real-world scenarios, such as professional sports, video game design, and drug-trafficking interdiction. In this work, we present Graph based Adversarial Modeling with Mutal Information (GrAMMI) for modeling the behavior of an adversarial opponent agent. GrAMMI is a novel graph neural network (GNN) based approach that uses mutual information maximization as an auxiliary objective to predict the current and future states of an adversarial opponent with partial observability. To evaluate GrAMMI, we design two large-scale, pursuit-evasion domains inspired by real-world scenarios, where a team of heterogeneous agents is tasked with tracking and interdicting a single adversarial agent, and the adversarial agent must evade detection while achieving its own objectives. With the mutual information formulation, GrAMMI outperforms all baselines in both domains and achieves 31.68% higher log-likelihood on average for future adversarial state predictions across both domains.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2023-07 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2828072210 |
source | Free E- Journals |
subjects | Computer & video games Domains Graph neural networks Modelling Pursuit-evasion games Tracking |
title | Learning Models of Adversarial Agent Behavior under Partial Observability |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-05T00%3A56%3A22IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Learning%20Models%20of%20Adversarial%20Agent%20Behavior%20under%20Partial%20Observability&rft.jtitle=arXiv.org&rft.au=Ye,%20Sean&rft.date=2023-07-05&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2828072210%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2828072210&rft_id=info:pmid/&rfr_iscdi=true |