Noisy GA resampling on evolved parameterized policies for stochastic constraint programming

Jing Tian, Tomohiro Murata

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

    Abstract

    Stochastic Constraint Programming is an extension of Constraint Programming for modeling and solving combinatorial problems which involve uncertainty in real world. Evolved Parameterized Policies (EPP) is the first incomplete approach to stochastic constraint problems which has higher performance rather than other methods, but still seems non-practical for large multi-stage problems due to scenarios exponentially growing. We proposed new resampling method called IDGAS based on Noisy GAs and other Evolutionary Computation algorithms, which aim to ensure the reliability while keeping in high search performance. In experiments on credit portfolio management with multi-stage, it performed more effective than conventional EPP and other resampling methods.

    Original languageEnglish
    Title of host publicationProceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012
    Pages1439-1442
    Number of pages4
    DOIs
    Publication statusPublished - 2012
    Event2012 International Conference on Computer Science and Service System, CSSS 2012 - Nanjing
    Duration: 2012 Aug 112012 Aug 13

    Other

    Other2012 International Conference on Computer Science and Service System, CSSS 2012
    CityNanjing
    Period12/8/1112/8/13

    Fingerprint

    Evolutionary algorithms
    Experiments
    Uncertainty

    Keywords

    • Evolved Parameterized Policies
    • Noisy GA
    • Resampling/Sampling
    • Stochastic Constraint Programming

    ASJC Scopus subject areas

    • Computer Science (miscellaneous)

    Cite this

    Tian, J., & Murata, T. (2012). Noisy GA resampling on evolved parameterized policies for stochastic constraint programming. In Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012 (pp. 1439-1442). [6394600] https://doi.org/10.1109/CSSS.2012.362

    Noisy GA resampling on evolved parameterized policies for stochastic constraint programming. / Tian, Jing; Murata, Tomohiro.

    Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012. 2012. p. 1439-1442 6394600.

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

    Tian, J & Murata, T 2012, Noisy GA resampling on evolved parameterized policies for stochastic constraint programming. in Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012., 6394600, pp. 1439-1442, 2012 International Conference on Computer Science and Service System, CSSS 2012, Nanjing, 12/8/11. https://doi.org/10.1109/CSSS.2012.362
    Tian J, Murata T. Noisy GA resampling on evolved parameterized policies for stochastic constraint programming. In Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012. 2012. p. 1439-1442. 6394600 https://doi.org/10.1109/CSSS.2012.362
    Tian, Jing ; Murata, Tomohiro. / Noisy GA resampling on evolved parameterized policies for stochastic constraint programming. Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012. 2012. pp. 1439-1442
    @inproceedings{fe6d31d52026411094fe08146658bae1,
    title = "Noisy GA resampling on evolved parameterized policies for stochastic constraint programming",
    abstract = "Stochastic Constraint Programming is an extension of Constraint Programming for modeling and solving combinatorial problems which involve uncertainty in real world. Evolved Parameterized Policies (EPP) is the first incomplete approach to stochastic constraint problems which has higher performance rather than other methods, but still seems non-practical for large multi-stage problems due to scenarios exponentially growing. We proposed new resampling method called IDGAS based on Noisy GAs and other Evolutionary Computation algorithms, which aim to ensure the reliability while keeping in high search performance. In experiments on credit portfolio management with multi-stage, it performed more effective than conventional EPP and other resampling methods.",
    keywords = "Evolved Parameterized Policies, Noisy GA, Resampling/Sampling, Stochastic Constraint Programming",
    author = "Jing Tian and Tomohiro Murata",
    year = "2012",
    doi = "10.1109/CSSS.2012.362",
    language = "English",
    isbn = "9780769547190",
    pages = "1439--1442",
    booktitle = "Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012",

    }

    TY - GEN

    T1 - Noisy GA resampling on evolved parameterized policies for stochastic constraint programming

    AU - Tian, Jing

    AU - Murata, Tomohiro

    PY - 2012

    Y1 - 2012

    N2 - Stochastic Constraint Programming is an extension of Constraint Programming for modeling and solving combinatorial problems which involve uncertainty in real world. Evolved Parameterized Policies (EPP) is the first incomplete approach to stochastic constraint problems which has higher performance rather than other methods, but still seems non-practical for large multi-stage problems due to scenarios exponentially growing. We proposed new resampling method called IDGAS based on Noisy GAs and other Evolutionary Computation algorithms, which aim to ensure the reliability while keeping in high search performance. In experiments on credit portfolio management with multi-stage, it performed more effective than conventional EPP and other resampling methods.

    AB - Stochastic Constraint Programming is an extension of Constraint Programming for modeling and solving combinatorial problems which involve uncertainty in real world. Evolved Parameterized Policies (EPP) is the first incomplete approach to stochastic constraint problems which has higher performance rather than other methods, but still seems non-practical for large multi-stage problems due to scenarios exponentially growing. We proposed new resampling method called IDGAS based on Noisy GAs and other Evolutionary Computation algorithms, which aim to ensure the reliability while keeping in high search performance. In experiments on credit portfolio management with multi-stage, it performed more effective than conventional EPP and other resampling methods.

    KW - Evolved Parameterized Policies

    KW - Noisy GA

    KW - Resampling/Sampling

    KW - Stochastic Constraint Programming

    UR - http://www.scopus.com/inward/record.url?scp=84873870567&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84873870567&partnerID=8YFLogxK

    U2 - 10.1109/CSSS.2012.362

    DO - 10.1109/CSSS.2012.362

    M3 - Conference contribution

    AN - SCOPUS:84873870567

    SN - 9780769547190

    SP - 1439

    EP - 1442

    BT - Proceedings - 2012 International Conference on Computer Science and Service System, CSSS 2012

    ER -