Constructive approximations to the q=1/2 maximum entropy distribution from redundant and noisy data

L. Rebollo-Neira*, A. Plastino

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    Abstract

    An approach adopted to consider the problem of constructing the q=1/2 maximum entropy distribution from redundant and noisy data was discussed. The advantage of this generalized approach, when dealing with very noisy data was illustrated by a numerical simulation. A strategy was proposed that evolved through different steps such as independent constraints were first preselected by recourse to a data independent technique. A backward approach was also proposed for reducing the parameters of such distributions. It was found that the sub-optimal strategies could be utilized in a broad range of situations.

    Original languageEnglish
    Article number021104
    JournalPhysical Review E
    Volume70
    Issue number2 1
    DOIs
    Publication statusPublished - 1 Aug 2004

    Bibliographical note

    ©2004 American Physical Society

    Fingerprint

    Dive into the research topics of 'Constructive approximations to the q=1/2 maximum entropy distribution from redundant and noisy data'. Together they form a unique fingerprint.

    Cite this