Emergence of conventions for efficiently resolving conflicts in complex networks

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    3 Citations (Scopus)

    Abstract

    We investigated the emergence of conventions for conflict resolutions in agent networks with various structures through pair wise reinforcement learning. Whereas coordinated agents encounter conflict situations in the course of actions, their resolutions are complex and computationally expensive due to mutual analysis of subsequent actions by both agents and communication costs of the interactions. Norms and conventions are expected to reduce these costs by regulating agent actions in recurrent conflicts. This paper describes a typical conflict situation using a Markov game and we investigated whether or not agents with a certain attitude to conflicts could learn the conventions of agent networks that had complex structures. We first examined the emergence of conventions and their characteristics in fully connected networks. Then, we compared them with the results from other agent network structures such as BA and CNN networks. We found the network structure strongly affected their emergence and the agents could sometimes learn no conventions although they could learn locally consistent actions for resolutions.

    Original languageEnglish
    Title of host publicationProceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages371-378
    Number of pages8
    Volume3
    ISBN (Print)9781479941438
    Publication statusPublished - 2014 Oct 16
    Event2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014 - Warsaw, Poland
    Duration: 2014 Aug 112014 Aug 14

    Other

    Other2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014
    CountryPoland
    CityWarsaw
    Period14/8/1114/8/14

    Fingerprint

    Complex networks
    Reinforcement learning
    Costs
    Communication

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Computer Networks and Communications
    • Software

    Cite this

    Sugawara, T. (2014). Emergence of conventions for efficiently resolving conflicts in complex networks. In Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014 (Vol. 3, pp. 371-378). Institute of Electrical and Electronics Engineers Inc..

    Emergence of conventions for efficiently resolving conflicts in complex networks. / Sugawara, Toshiharu.

    Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014. Vol. 3 Institute of Electrical and Electronics Engineers Inc., 2014. p. 371-378.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Sugawara, T 2014, Emergence of conventions for efficiently resolving conflicts in complex networks. in Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014. vol. 3, Institute of Electrical and Electronics Engineers Inc., pp. 371-378, 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014, Warsaw, Poland, 14/8/11.
    Sugawara T. Emergence of conventions for efficiently resolving conflicts in complex networks. In Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014. Vol. 3. Institute of Electrical and Electronics Engineers Inc. 2014. p. 371-378
    Sugawara, Toshiharu. / Emergence of conventions for efficiently resolving conflicts in complex networks. Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014. Vol. 3 Institute of Electrical and Electronics Engineers Inc., 2014. pp. 371-378
    @inproceedings{47acdf0d88f3424894fcdd1bacfe6c4d,
    title = "Emergence of conventions for efficiently resolving conflicts in complex networks",
    abstract = "We investigated the emergence of conventions for conflict resolutions in agent networks with various structures through pair wise reinforcement learning. Whereas coordinated agents encounter conflict situations in the course of actions, their resolutions are complex and computationally expensive due to mutual analysis of subsequent actions by both agents and communication costs of the interactions. Norms and conventions are expected to reduce these costs by regulating agent actions in recurrent conflicts. This paper describes a typical conflict situation using a Markov game and we investigated whether or not agents with a certain attitude to conflicts could learn the conventions of agent networks that had complex structures. We first examined the emergence of conventions and their characteristics in fully connected networks. Then, we compared them with the results from other agent network structures such as BA and CNN networks. We found the network structure strongly affected their emergence and the agents could sometimes learn no conventions although they could learn locally consistent actions for resolutions.",
    author = "Toshiharu Sugawara",
    year = "2014",
    month = "10",
    day = "16",
    language = "English",
    isbn = "9781479941438",
    volume = "3",
    pages = "371--378",
    booktitle = "Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - GEN

    T1 - Emergence of conventions for efficiently resolving conflicts in complex networks

    AU - Sugawara, Toshiharu

    PY - 2014/10/16

    Y1 - 2014/10/16

    N2 - We investigated the emergence of conventions for conflict resolutions in agent networks with various structures through pair wise reinforcement learning. Whereas coordinated agents encounter conflict situations in the course of actions, their resolutions are complex and computationally expensive due to mutual analysis of subsequent actions by both agents and communication costs of the interactions. Norms and conventions are expected to reduce these costs by regulating agent actions in recurrent conflicts. This paper describes a typical conflict situation using a Markov game and we investigated whether or not agents with a certain attitude to conflicts could learn the conventions of agent networks that had complex structures. We first examined the emergence of conventions and their characteristics in fully connected networks. Then, we compared them with the results from other agent network structures such as BA and CNN networks. We found the network structure strongly affected their emergence and the agents could sometimes learn no conventions although they could learn locally consistent actions for resolutions.

    AB - We investigated the emergence of conventions for conflict resolutions in agent networks with various structures through pair wise reinforcement learning. Whereas coordinated agents encounter conflict situations in the course of actions, their resolutions are complex and computationally expensive due to mutual analysis of subsequent actions by both agents and communication costs of the interactions. Norms and conventions are expected to reduce these costs by regulating agent actions in recurrent conflicts. This paper describes a typical conflict situation using a Markov game and we investigated whether or not agents with a certain attitude to conflicts could learn the conventions of agent networks that had complex structures. We first examined the emergence of conventions and their characteristics in fully connected networks. Then, we compared them with the results from other agent network structures such as BA and CNN networks. We found the network structure strongly affected their emergence and the agents could sometimes learn no conventions although they could learn locally consistent actions for resolutions.

    UR - http://www.scopus.com/inward/record.url?scp=84931679276&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84931679276&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:84931679276

    SN - 9781479941438

    VL - 3

    SP - 371

    EP - 378

    BT - Proceedings - 2014 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Workshops, WI-IAT 2014

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -