Elsevier

Information Sciences

Volume 7, 1974, Pages 237-252
Information Sciences

The implicit conditioning method in statistical mechanics

https://doi.org/10.1016/0020-0255(74)90016-4Get rights and content

Abstract

It is well known that least mean square estimation can be employed to calculate conditional means, a procedure called the implicit conditioning method in this paper. It is possible to construct a priori probability densities of tractable form that, when conditioned on certain sets of variables, reduce to conditional probability densities which are identical to the canonical probability densities occurring in the statistical mechanics of certain classical systems. This yields a new variational principle for the calculation of canonical mean values in classical statistical mechanics. In this paper, two versions of this variational principle are applied to a simple lattice system to yield approximate expressions for the canonical mean values of certain properties of physical interest.

References (13)

  • E.H. Lieb et al.

    Mathematical Physics in One Dimension

    (1966)
  • D.G. Lainiotis

    Optimal linear smoothing

    Int. J. Control

    (1973)
    D.G. Lainiotis

    Optimal nonlinear estimation

    Int. J. Control

    (1971)
    D.G. Lainiotis

    Optimal adaptive estimation: Structural and parameter adaption

    IEEE Trans. Auto. Control

    (1971)
  • J.M. Richardson et al.

    A decisiontheoretical approach to the detection of ECG abnormalities: II. Approximate treatment of the detection of ventricular extrasystoles

    Math. Biosci.

    (1971)
  • J.M. Richardson

    Pattern recognition and statistical mechanics

    J. Stat. Phys.

    (1969)
There are more references available in the full text version of this article.

Cited by (0)

View full text