SSDPOP: Improving the privacy of DCOP with secret sharing

Rachel Greenstadt, Barbara Grosz, Michael D. Smith

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

    Abstract

    Multi-agent systems designed to work collaboratively with groups of people typically require private information that people will entrust to them only if they have assurance that this information will be protected. Although Distributed Constraint Optimization (DCOP) has emerged as a prominent technique for multiagent coordination, existing algorithms for solving DCOP problems do not adeqately protect agents' privacy. This paper analyzes privacy protection and loss in existing DCOP algorithms. It presents a new algorithm, SSDPOP, which augments a prominent DCOP algorithm (DPOP) with secret sharing techniques. This approach significantly reduces privacy loss, while preserving the structure of the DPOP algorithm and introducing only minimal computational overhead. Results show that SSDPOP reduces privacy loss by 29 - 88% on average over DPOP.

    Original languageEnglish (US)
    Title of host publicationAAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems
    Pages1098-1100
    Number of pages3
    DOIs
    StatePublished - Dec 1 2007
    Event6th International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS'07 - Honolulu, HI, United States
    Duration: May 14 2008May 18 2008

    Publication series

    NameProceedings of the International Conference on Autonomous Agents

    Conference

    Conference6th International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS'07
    CountryUnited States
    CityHonolulu, HI
    Period5/14/085/18/08

    Fingerprint

    Secret Sharing
    Privacy
    Optimization
    Optimization Algorithm
    Privacy Protection
    Private Information
    Multi-agent Systems
    Multi agent systems
    Optimization Problem

    Keywords

    • Constraint reasoning
    • DCOP
    • Privacy

    ASJC Scopus subject areas

    • Software
    • Artificial Intelligence
    • Computer Networks and Communications
    • Theoretical Computer Science

    Cite this

    Greenstadt, R., Grosz, B., & Smith, M. D. (2007). SSDPOP: Improving the privacy of DCOP with secret sharing. In AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems (pp. 1098-1100). [171] (Proceedings of the International Conference on Autonomous Agents). https://doi.org/10.1145/1329125.1329333

    SSDPOP : Improving the privacy of DCOP with secret sharing. / Greenstadt, Rachel; Grosz, Barbara; Smith, Michael D.

    AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems. 2007. p. 1098-1100 171 (Proceedings of the International Conference on Autonomous Agents).

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

    Greenstadt, R, Grosz, B & Smith, MD 2007, SSDPOP: Improving the privacy of DCOP with secret sharing. in AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems., 171, Proceedings of the International Conference on Autonomous Agents, pp. 1098-1100, 6th International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS'07, Honolulu, HI, United States, 5/14/08. https://doi.org/10.1145/1329125.1329333
    Greenstadt R, Grosz B, Smith MD. SSDPOP: Improving the privacy of DCOP with secret sharing. In AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems. 2007. p. 1098-1100. 171. (Proceedings of the International Conference on Autonomous Agents). https://doi.org/10.1145/1329125.1329333
    Greenstadt, Rachel ; Grosz, Barbara ; Smith, Michael D. / SSDPOP : Improving the privacy of DCOP with secret sharing. AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems. 2007. pp. 1098-1100 (Proceedings of the International Conference on Autonomous Agents).
    @inproceedings{e84310307bc04356ada0142a873a2932,
    title = "SSDPOP: Improving the privacy of DCOP with secret sharing",
    abstract = "Multi-agent systems designed to work collaboratively with groups of people typically require private information that people will entrust to them only if they have assurance that this information will be protected. Although Distributed Constraint Optimization (DCOP) has emerged as a prominent technique for multiagent coordination, existing algorithms for solving DCOP problems do not adeqately protect agents' privacy. This paper analyzes privacy protection and loss in existing DCOP algorithms. It presents a new algorithm, SSDPOP, which augments a prominent DCOP algorithm (DPOP) with secret sharing techniques. This approach significantly reduces privacy loss, while preserving the structure of the DPOP algorithm and introducing only minimal computational overhead. Results show that SSDPOP reduces privacy loss by 29 - 88{\%} on average over DPOP.",
    keywords = "Constraint reasoning, DCOP, Privacy",
    author = "Rachel Greenstadt and Barbara Grosz and Smith, {Michael D.}",
    year = "2007",
    month = "12",
    day = "1",
    doi = "10.1145/1329125.1329333",
    language = "English (US)",
    isbn = "9788190426275",
    series = "Proceedings of the International Conference on Autonomous Agents",
    pages = "1098--1100",
    booktitle = "AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems",

    }

    TY - GEN

    T1 - SSDPOP

    T2 - Improving the privacy of DCOP with secret sharing

    AU - Greenstadt, Rachel

    AU - Grosz, Barbara

    AU - Smith, Michael D.

    PY - 2007/12/1

    Y1 - 2007/12/1

    N2 - Multi-agent systems designed to work collaboratively with groups of people typically require private information that people will entrust to them only if they have assurance that this information will be protected. Although Distributed Constraint Optimization (DCOP) has emerged as a prominent technique for multiagent coordination, existing algorithms for solving DCOP problems do not adeqately protect agents' privacy. This paper analyzes privacy protection and loss in existing DCOP algorithms. It presents a new algorithm, SSDPOP, which augments a prominent DCOP algorithm (DPOP) with secret sharing techniques. This approach significantly reduces privacy loss, while preserving the structure of the DPOP algorithm and introducing only minimal computational overhead. Results show that SSDPOP reduces privacy loss by 29 - 88% on average over DPOP.

    AB - Multi-agent systems designed to work collaboratively with groups of people typically require private information that people will entrust to them only if they have assurance that this information will be protected. Although Distributed Constraint Optimization (DCOP) has emerged as a prominent technique for multiagent coordination, existing algorithms for solving DCOP problems do not adeqately protect agents' privacy. This paper analyzes privacy protection and loss in existing DCOP algorithms. It presents a new algorithm, SSDPOP, which augments a prominent DCOP algorithm (DPOP) with secret sharing techniques. This approach significantly reduces privacy loss, while preserving the structure of the DPOP algorithm and introducing only minimal computational overhead. Results show that SSDPOP reduces privacy loss by 29 - 88% on average over DPOP.

    KW - Constraint reasoning

    KW - DCOP

    KW - Privacy

    UR - http://www.scopus.com/inward/record.url?scp=60349086001&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=60349086001&partnerID=8YFLogxK

    U2 - 10.1145/1329125.1329333

    DO - 10.1145/1329125.1329333

    M3 - Conference contribution

    AN - SCOPUS:60349086001

    SN - 9788190426275

    T3 - Proceedings of the International Conference on Autonomous Agents

    SP - 1098

    EP - 1100

    BT - AAMAS'07 - Proceedings of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems

    ER -