Statistical physics of pairwise probability models

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Statistical physics of pairwise probability models. / Roudi, Yasser; Aurell, Erik; Hertz, John.

In: Frontiers in Computational Neuroscience, Vol. 3, No. 22, 2009, p. (art 22).

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Roudi, Y, Aurell, E & Hertz, J 2009, 'Statistical physics of pairwise probability models', Frontiers in Computational Neuroscience, vol. 3, no. 22, pp. (art 22). https://doi.org/10.3389/neuro.10.022.2009

APA

Roudi, Y., Aurell, E., & Hertz, J. (2009). Statistical physics of pairwise probability models. Frontiers in Computational Neuroscience, 3(22), (art 22). https://doi.org/10.3389/neuro.10.022.2009

Vancouver

Roudi Y, Aurell E, Hertz J. Statistical physics of pairwise probability models. Frontiers in Computational Neuroscience. 2009;3(22):(art 22). https://doi.org/10.3389/neuro.10.022.2009

Author

Roudi, Yasser ; Aurell, Erik ; Hertz, John. / Statistical physics of pairwise probability models. In: Frontiers in Computational Neuroscience. 2009 ; Vol. 3, No. 22. pp. (art 22).

Bibtex

@article{bf887d700cc811df825d000ea68e967b,
title = "Statistical physics of pairwise probability models",
abstract = "(dansk abstrakt findes ikke)Statistical models for describing the probability distribution over the states of biological systems are commonly used for dimensional reduction. Among these models, pairwise models are very attractive in part because they can be fit using a reasonable amount of  data: knowledge of the means and correlations between pairs of elements in the system is sufficient. Not surprisingly, then, using pairwise models for studying neural data has been the focus of many studies in recent years. In this paper, we describe how tools from statistical physics can be employed for studying and using pairwise models. We build on our previous work on the subject and study the relation between different methods for fitting these models and evaluating their quality. In particular, using data from simulated cortical networks we study how the quality of various approximate methods for inferring the parameters in a pairwise modeldepends on the time bin chosen for binning the data. We also study the effect of the size of the time bin on the model quality itself, again using simulated data. We show that using finer time bins increases thequality of the pairwise model. We offer new ways of deriving the expressions reported in our previous work for assessing the quality of pairwise models. Udgivelsesdato: 17 Nov.",
author = "Yasser Roudi and Erik Aurell and John Hertz",
note = "Paper id:: DOI: 10.3389/neuro.10.022.2009",
year = "2009",
doi = "10.3389/neuro.10.022.2009",
language = "English",
volume = "3",
pages = "(art 22)",
journal = "Frontiers in Computational Neuroscience",
issn = "1662-5188",
publisher = "Frontiers Research Foundation",
number = "22",

}

RIS

TY - JOUR

T1 - Statistical physics of pairwise probability models

AU - Roudi, Yasser

AU - Aurell, Erik

AU - Hertz, John

N1 - Paper id:: DOI: 10.3389/neuro.10.022.2009

PY - 2009

Y1 - 2009

N2 - (dansk abstrakt findes ikke)Statistical models for describing the probability distribution over the states of biological systems are commonly used for dimensional reduction. Among these models, pairwise models are very attractive in part because they can be fit using a reasonable amount of  data: knowledge of the means and correlations between pairs of elements in the system is sufficient. Not surprisingly, then, using pairwise models for studying neural data has been the focus of many studies in recent years. In this paper, we describe how tools from statistical physics can be employed for studying and using pairwise models. We build on our previous work on the subject and study the relation between different methods for fitting these models and evaluating their quality. In particular, using data from simulated cortical networks we study how the quality of various approximate methods for inferring the parameters in a pairwise modeldepends on the time bin chosen for binning the data. We also study the effect of the size of the time bin on the model quality itself, again using simulated data. We show that using finer time bins increases thequality of the pairwise model. We offer new ways of deriving the expressions reported in our previous work for assessing the quality of pairwise models. Udgivelsesdato: 17 Nov.

AB - (dansk abstrakt findes ikke)Statistical models for describing the probability distribution over the states of biological systems are commonly used for dimensional reduction. Among these models, pairwise models are very attractive in part because they can be fit using a reasonable amount of  data: knowledge of the means and correlations between pairs of elements in the system is sufficient. Not surprisingly, then, using pairwise models for studying neural data has been the focus of many studies in recent years. In this paper, we describe how tools from statistical physics can be employed for studying and using pairwise models. We build on our previous work on the subject and study the relation between different methods for fitting these models and evaluating their quality. In particular, using data from simulated cortical networks we study how the quality of various approximate methods for inferring the parameters in a pairwise modeldepends on the time bin chosen for binning the data. We also study the effect of the size of the time bin on the model quality itself, again using simulated data. We show that using finer time bins increases thequality of the pairwise model. We offer new ways of deriving the expressions reported in our previous work for assessing the quality of pairwise models. Udgivelsesdato: 17 Nov.

U2 - 10.3389/neuro.10.022.2009

DO - 10.3389/neuro.10.022.2009

M3 - Journal article

C2 - 19949460

VL - 3

SP - (art 22)

JO - Frontiers in Computational Neuroscience

JF - Frontiers in Computational Neuroscience

SN - 1662-5188

IS - 22

ER -

ID: 17272924