Combining spatially distributed predictions from neural networks

Christopher K. I. Williams

    Research output: Working paperTechnical report

    Abstract

    In this report we discuss the problem of combining spatially-distributed predictions from neural networks. An example of this problem is the prediction of a wind vector-field from remote-sensing data by combining bottom-up predictions (wind vector predictions on a pixel-by-pixel basis) with prior knowledge about wind-field configurations. This task can be achieved using the scaled-likelihood method, which has been used by Morgan and Bourlard (1995) and Smyth (1994), in the context of Hidden Markov modelling
    Original languageEnglish
    Place of PublicationBirmingham B4 7ET, UK
    PublisherAston University
    Number of pages4
    ISBN (Print)NCRG/97/026
    Publication statusPublished - 1997

    Fingerprint

    Neural networks
    Pixels
    Remote sensing

    Keywords

    • spatially-distributed
    • neural network
    • Hidden Markov modelling

    Cite this

    Williams, C. K. I. (1997). Combining spatially distributed predictions from neural networks. Birmingham B4 7ET, UK: Aston University.
    Williams, Christopher K. I. / Combining spatially distributed predictions from neural networks. Birmingham B4 7ET, UK : Aston University, 1997.
    @techreport{164a13cd5e8543fd9d84060a046d16da,
    title = "Combining spatially distributed predictions from neural networks",
    abstract = "In this report we discuss the problem of combining spatially-distributed predictions from neural networks. An example of this problem is the prediction of a wind vector-field from remote-sensing data by combining bottom-up predictions (wind vector predictions on a pixel-by-pixel basis) with prior knowledge about wind-field configurations. This task can be achieved using the scaled-likelihood method, which has been used by Morgan and Bourlard (1995) and Smyth (1994), in the context of Hidden Markov modelling",
    keywords = "spatially-distributed, neural network, Hidden Markov modelling",
    author = "Williams, {Christopher K. I.}",
    year = "1997",
    language = "English",
    isbn = "NCRG/97/026",
    publisher = "Aston University",
    type = "WorkingPaper",
    institution = "Aston University",

    }

    Williams, CKI 1997 'Combining spatially distributed predictions from neural networks' Aston University, Birmingham B4 7ET, UK.

    Combining spatially distributed predictions from neural networks. / Williams, Christopher K. I.

    Birmingham B4 7ET, UK : Aston University, 1997.

    Research output: Working paperTechnical report

    TY - UNPB

    T1 - Combining spatially distributed predictions from neural networks

    AU - Williams, Christopher K. I.

    PY - 1997

    Y1 - 1997

    N2 - In this report we discuss the problem of combining spatially-distributed predictions from neural networks. An example of this problem is the prediction of a wind vector-field from remote-sensing data by combining bottom-up predictions (wind vector predictions on a pixel-by-pixel basis) with prior knowledge about wind-field configurations. This task can be achieved using the scaled-likelihood method, which has been used by Morgan and Bourlard (1995) and Smyth (1994), in the context of Hidden Markov modelling

    AB - In this report we discuss the problem of combining spatially-distributed predictions from neural networks. An example of this problem is the prediction of a wind vector-field from remote-sensing data by combining bottom-up predictions (wind vector predictions on a pixel-by-pixel basis) with prior knowledge about wind-field configurations. This task can be achieved using the scaled-likelihood method, which has been used by Morgan and Bourlard (1995) and Smyth (1994), in the context of Hidden Markov modelling

    KW - spatially-distributed

    KW - neural network

    KW - Hidden Markov modelling

    M3 - Technical report

    SN - NCRG/97/026

    BT - Combining spatially distributed predictions from neural networks

    PB - Aston University

    CY - Birmingham B4 7ET, UK

    ER -

    Williams CKI. Combining spatially distributed predictions from neural networks. Birmingham B4 7ET, UK: Aston University. 1997.