Time series prediction via neural network inversion

    Research output: Contribution to journalArticle

    1 Citation (Scopus)

    Abstract

    In this work, we propose neural network inversion of a backward predictor as a technique for multi-step prediction of dynamic time series. It may be difficult to train a large network to capture the correlation that exists in some dynamic time series represented by small data sets. The new approach combines an estimate obtained from a forward predictor with an estimate obtained by inverting a backward predictor to more efficiently capture the correlation and to achieve more accurate predictions. Inversion allows us to make causal use of prediction backward in time. Also a new regularization method is developed to make neural network inversion less ill-posed. Experimental results on two benchmark time series demonstrate the new approach's significant improvement over standard forward prediction, given comparable complexity.

    Original languageEnglish (US)
    Pages (from-to)1049-1052
    Number of pages4
    JournalProceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
    Volume2
    StatePublished - 1999

    Fingerprint

    Time series
    inversions
    Neural networks
    predictions
    estimates

    All Science Journal Classification (ASJC) codes

    • Signal Processing
    • Electrical and Electronic Engineering
    • Acoustics and Ultrasonics

    Cite this

    @article{9b4027cea8dd4ce3b9c47668929fb2e0,
    title = "Time series prediction via neural network inversion",
    abstract = "In this work, we propose neural network inversion of a backward predictor as a technique for multi-step prediction of dynamic time series. It may be difficult to train a large network to capture the correlation that exists in some dynamic time series represented by small data sets. The new approach combines an estimate obtained from a forward predictor with an estimate obtained by inverting a backward predictor to more efficiently capture the correlation and to achieve more accurate predictions. Inversion allows us to make causal use of prediction backward in time. Also a new regularization method is developed to make neural network inversion less ill-posed. Experimental results on two benchmark time series demonstrate the new approach's significant improvement over standard forward prediction, given comparable complexity.",
    author = "Lian Yan and Miller, {David Jonathan}",
    year = "1999",
    language = "English (US)",
    volume = "2",
    pages = "1049--1052",
    journal = "Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing",
    issn = "0736-7791",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",

    }

    TY - JOUR

    T1 - Time series prediction via neural network inversion

    AU - Yan, Lian

    AU - Miller, David Jonathan

    PY - 1999

    Y1 - 1999

    N2 - In this work, we propose neural network inversion of a backward predictor as a technique for multi-step prediction of dynamic time series. It may be difficult to train a large network to capture the correlation that exists in some dynamic time series represented by small data sets. The new approach combines an estimate obtained from a forward predictor with an estimate obtained by inverting a backward predictor to more efficiently capture the correlation and to achieve more accurate predictions. Inversion allows us to make causal use of prediction backward in time. Also a new regularization method is developed to make neural network inversion less ill-posed. Experimental results on two benchmark time series demonstrate the new approach's significant improvement over standard forward prediction, given comparable complexity.

    AB - In this work, we propose neural network inversion of a backward predictor as a technique for multi-step prediction of dynamic time series. It may be difficult to train a large network to capture the correlation that exists in some dynamic time series represented by small data sets. The new approach combines an estimate obtained from a forward predictor with an estimate obtained by inverting a backward predictor to more efficiently capture the correlation and to achieve more accurate predictions. Inversion allows us to make causal use of prediction backward in time. Also a new regularization method is developed to make neural network inversion less ill-posed. Experimental results on two benchmark time series demonstrate the new approach's significant improvement over standard forward prediction, given comparable complexity.

    UR - http://www.scopus.com/inward/record.url?scp=0032627023&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=0032627023&partnerID=8YFLogxK

    M3 - Article

    VL - 2

    SP - 1049

    EP - 1052

    JO - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

    JF - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

    SN - 0736-7791

    ER -