Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 9 of 9 for “"Expectation Propagation"”.

  1. Extending expectation propagation for graphical models

    … networks. This thesis proposes extensions of expectation propagation, a powerful generalization of loopy belief propagation, to develop efficient Bayesian inference and learning algorithms for graphical models. The first two chapters of the thesis present inference algorithms for generative …

    mit Repository record for Extending expectation propagation for graphical models (opens in a new tab)

  2. Probabilistic Machine Learning for Circular Statistics: Models and inference using the Multivariate Generalised von Mises distribution

    … we investigate the use of Variational Inference, Expectation Propagation and Markov chain Monte Carlo methods. The variational inference route taken was a mean field approach to efficiently leverage the mGvM tractable conditionals and create a baseline for comparison with other methods. Then, an …

    cambridge Repository record for Probabilistic Machine Learning for Circular Statistics: Models and inference using the Multivariate Generalised von Mises distribution (opens in a new tab)

  3. Multi-Way Block Models

    … sampling, collapsed variational Bayesian, and expectation propagation approaches. Comparative simulation studies show that the four implementation algorithms achieve meaningful parameter estimates for the latent membership and block structure from correlation and network among the subjects.

    ohiolink Repository record for Multi-Way Block Models (opens in a new tab)

  4. Robust Prediction of Large Spatio-Temporal Datasets

    … inference algorithm in the framework of Expectation Propagation (EP). Extensive experimental evaluations, based on both simulation and real-life data sets, demonstrated the robustness and the efficiency of our Student-t prediction model compared with the STRE model.

    vt Repository record for Robust Prediction of Large Spatio-Temporal Datasets (opens in a new tab)

  5. Efficient Deterministic Approximate Bayesian Inference for Gaussian Process models

    … posterior approximation framework based on Power-Expectation Propagation for Gaussian process regression and classification. This framework relies on a structured approximate Gaussian process posterior based on a small number of pseudo-points, which is judiciously chosen to summarise the actual …

    cambridge Repository record for Efficient Deterministic Approximate Bayesian Inference for Gaussian Process models (opens in a new tab)

  6. Estimation of GMRFs by recursive cavity modeling

    … projection filtering, Markov-blanket filtering, expectation propagation and other methods relying on reduction of embedded models. These connections are explored and important distinctions and extensions are noted. The author believes this thesis represents a significant generalization of …

    mit Repository record for Estimation of GMRFs by recursive cavity modeling (opens in a new tab)

  7. Approximate Inference: New Visions

    … by drawing inspiration from the well known expectation propagation and message passing algorithms. Both approaches provide a unifying view of existing variational methods from different algorithmic perspectives. We also demonstrate that they lead to better calibrated inference results for …

    cambridge Repository record for Approximate Inference: New Visions (opens in a new tab)

  8. Natural gradient methods in statistics and machine learning

    … a novel natural-gradient interpretation of the expectation propagation (EP) algorithm of Minka (2001) to motivate two new natural-gradient-based EP variants that have particularly desirable properties in black-box inference settings. Black-box inference methods allow practitioners to answer …

    cambridge Repository record for Natural gradient methods in statistics and machine learning (opens in a new tab)

  9. Low dimensional visualization and modelling of data using distance-based models

    … that LSMs can instead be estimated with expectation propagation (EP). Necessary changes to the algorithm and the optimization procedures are shown and the implications of pairwise likelihood sites are thorougly discussed. Comparing EP to other methods of estimation for LSMs (MCMC and …

    tu-berlin Repository record for Low dimensional visualization and modelling of data using distance-based models (opens in a new tab)