Academic
Publications
An Introduction to Probabilistic Graphical Models

An Introduction to Probabilistic Graphical Models,Michael I. Jordan

An Introduction to Probabilistic Graphical Models   (Citations: 78)
BibTex | RIS | RefWorks Download
Published in 2003.
Cumulative Annual
    • ...the BP algorithm can be found at [28] and [29]...

    Ying Huet al. Real-time state estimation on micro-grids

    • ...First, we use the demonstration trajectories y as observations to the (unobserved) reference trajectory z in a Kalman smoother [28], [17], which will produce (Gaussian) distributions of the states along the reference trajectory z. The Kalman smoother is used as part of the EM-algorithm [13], [17], which iteratively and alternatingly infers the distributions of the reference trajectory given the current model parameters, and updates the ...
    • ...First, we use the demonstration trajectories y as observations to the (unobserved) reference trajectory z in a Kalman smoother [28], [17], which will produce (Gaussian) distributions of the states along the reference trajectory z. The Kalman smoother is used as part of the EM-algorithm [13], [17], which iteratively and alternatingly infers the distributions of the reference trajectory given the current model parameters, and updates the ...
    • ...We will not describe the Kalman smoother in full detail here, but refer to [17] for an excellent treatise...
    • ...As the “log” is eliminated against the “exp” in the Gaussian probability-density function, the expectations can be brought inside, and a closed form update rule for R (with fixed τ) can be derived (see [17] for more details)...

    Jur van den Berget al. Superhuman performance of surgical tasks by robots using iterative lea...

    • ...assumption. That is, subsequent samples are no longer assumed independent [9], [10]...

    Daniel M. Steinberget al. Towards autonomous habitat classification using Gaussian Mixture Model...

    • ...Hence, EM algorithm can be used to estimate the un­ known parameters of this model, namely (7fo , /10 , �o) and (7f1' /1 1, �d [10] (chapters 10 and 11)...
    • ...This problem can be solved recur­ sively in a neat way and in the mean time EM algorithm can be used to derive the mixture components [10]...

    Nima Noorshamset al. Centralized and decentralized cooperative spectrum sensing in cognitiv...

    • ...A Probabilistic Graphical Model is a topologic structure, in which the nodes represent the states and the link represent the probability of the dependency [1] [2]...
    • ...Belief Propagation methods are based on the so called sum-product algorithm [2]...
    • ...The two models can change to each other by factor graph [1] [2]...
    • ...The equation (4), (5) and (6) form the sequential belief propagation algorithm [2]...

    Lin Zhenget al. Articulated Body tracking based on sequential belief propagation

Sort by: