The Experts below are selected from a list of 284982 Experts worldwide ranked by ideXlab platform

Zoubin Ghahramani - One of the best experts on this subject based on the ideXlab platform.

  • latent gaussian Processes for distribution estimation of multivariate categorical data
    arXiv: Machine Learning, 2015
    Co-Authors: Yarin Gal, Yutian Chen, Zoubin Ghahramani
    Abstract:

    Multivariate categorical data occur in many applications of machine learning. One of the main difficulties with these vectors of categorical variables is sparsity. The number of possible observations grows exponentially with vector length, but dataset diversity might be poor in comparison. Recent models have gained significant improvement in supervised tasks with this data. These models embed observations in a continuous space to capture similarities between them. Building on these ideas we propose a Bayesian model for the unsupervised task of distribution estimation of multivariate categorical data. We model vectors of categorical variables as generated from a non-linear transformation of a continuous latent space. Non-linearity captures multi-modality in the distribution. The continuous representation addresses sparsity. Our model ties together many existing models, linking the linear categorical latent Gaussian model, the Gaussian Process latent variable model, and Gaussian Process Classification. We derive inference for our model based on recent developments in sampling based variational inference. We show empirically that the model outperforms its linear and discrete counterparts in imputation tasks of sparse data.

  • scalable variational gaussian Process Classification
    arXiv: Machine Learning, 2014
    Co-Authors: James Hensman, Alexander G. De G. Matthews, Zoubin Ghahramani
    Abstract:

    Gaussian Process Classification is a popular method with a number of appealing properties. We show how to scale the model within a variational inducing point framework, outperforming the state of the art on benchmark datasets. Importantly, the variational formulation can be exploited to allow Classification in problems with millions of data points, as we demonstrate in experiments.

  • bayesian gaussian Process Classification with the em ep algorithm
    IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006
    Co-Authors: Hyunchul Kim, Zoubin Ghahramani
    Abstract:

    Gaussian Process classifiers (GPCs) are Bayesian probabilistic kernel classifiers. In GPCs, the probability of belonging to a certain class at an input location is monotonically related to the value of some latent function at that location. Starting from a Gaussian Process prior over this latent function, data are used to infer both the posterior over the latent function and the values of hyperparameters to determine various aspects of the function. Recently, the expectation propagation (EP) approach has been proposed to infer the posterior over the latent function. Based on this work, we present an approximate EM algorithm, the EM-EP algorithm, to learn both the latent function and the hyperparameters. This algorithm is found to converge in practice and provides an efficient Bayesian framework for learning hyperparameters of the kernel. A multiclass extension of the EM-EP algorithm for GPCs is also derived. In the experimental results, the EM-EP algorithms are as good or better than other methods for GPCs or support vector machines (SVMs) with cross-validation

Daniel Hernandezlobato - One of the best experts on this subject based on the ideXlab platform.

  • scalable multi class gaussian Process Classification using expectation propagation
    arXiv: Machine Learning, 2017
    Co-Authors: Carlos Villacampacalvo, Daniel Hernandezlobato
    Abstract:

    This paper describes an expectation propagation (EP) method for multi-class Classification with Gaussian Processes that scales well to very large datasets. In such a method the estimate of the log-marginal-likelihood involves a sum across the data instances. This enables efficient training using stochastic gradients and mini-batches. When this type of training is used, the computational cost does not depend on the number of data instances $N$. Furthermore, extra assumptions in the approximate inference Process make the memory cost independent of $N$. The consequence is that the proposed EP method can be used on datasets with millions of instances. We compare empirically this method with alternative approaches that approximate the required computations using variational inference. The results show that it performs similar or even better than these techniques, which sometimes give significantly worse predictive distributions in terms of the test log-likelihood. Besides this, the training Process of the proposed approach also seems to converge in a smaller number of iterations.

  • scalable gaussian Process Classification via expectation propagation
    International Conference on Artificial Intelligence and Statistics, 2016
    Co-Authors: Daniel Hernandezlobato, Jose Miguel Hernandezlobato
    Abstract:

    Variational methods have been recently considered for scaling the training Process of Gaussian Process classifiers to large datasets. As an alternative, we describe here how to train these classifiers efficiently using expectation propagation (EP). The proposed EP method allows to train Gaussian Process classifiers on very large datasets, with millions of instances, that were out of the reach of previous implementations of EP. More precisely, it can be used for (i) training in a distributed fashion where the data instances are sent to different nodes in which the required computations are carried out, and for (ii) maximizing an estimate of the marginal likelihood using a stochastic approximation of the gradient. Several experiments involving large datasets show that the method described is competitive with the variational approach.

  • scalable gaussian Process Classification via expectation propagation
    International Conference on Artificial Intelligence and Statistics, 2016
    Co-Authors: Daniel Hernandezlobato, Jose Miguel Hernandezlobato
    Abstract:

    Copyright 2016 by the authors. Variational methods have been recently considered for scaling the training Process of Gaussian Process classifiers to large datasets. As an alternative, we describe here how to train these classifiers efficiently using expectation propagation (EP). The proposed EP method allows to train Gaussian Process classifiers on very large datasets, with millions of instances, that were out of the reach of previous implementations of EP. More precisely, it can be used for (i) training in a distributed fashion where the data instances are sent to different nodes in which the required computations are carried out, and for (ii) maximizing an estimate of the marginal likelihood using a stochastic approximation of the gradient. Several experiments involving large datasets show that the method described is competitive with the variational approach.

  • scalable gaussian Process Classification via expectation propagation
    arXiv: Machine Learning, 2015
    Co-Authors: Daniel Hernandezlobato, Jose Miguel Hernandezlobato
    Abstract:

    Variational methods have been recently considered for scaling the training Process of Gaussian Process classifiers to large datasets. As an alternative, we describe here how to train these classifiers efficiently using expectation propagation. The proposed method allows for handling datasets with millions of data instances. More precisely, it can be used for (i) training in a distributed fashion where the data instances are sent to different nodes in which the required computations are carried out, and for (ii) maximizing an estimate of the marginal likelihood using a stochastic approximation of the gradient. Several experiments indicate that the method described is competitive with the variational approach.

Manfred Opper - One of the best experts on this subject based on the ideXlab platform.

  • efficient gaussian Process Classification using polya gamma data augmentation
    National Conference on Artificial Intelligence, 2019
    Co-Authors: Florian Wenzel, Theo Galyfajou, Christian Donner, Marius Kloft, Manfred Opper
    Abstract:

    We propose a scalable stochastic variational approach to GP Classification building on Polya-Gamma data augmentation and inducing points. Unlike former approaches, we obtain closed-form updates based on natural gradients that lead to efficient optimization. We evaluate the algorithm on real-world datasets containing up to 11 million data points and demonstrate that it is up to two orders of magnitude faster than the state-of-the-art while being competitive in terms of prediction performance.

  • efficient gaussian Process Classification using polya gamma data augmentation
    arXiv: Machine Learning, 2018
    Co-Authors: Florian Wenzel, Theo Galyfajou, Christian Donner, Marius Kloft, Manfred Opper
    Abstract:

    We propose an efficient stochastic variational approach to GP Classification building on Polya- Gamma data augmentation and inducing points, which is based on closed-form updates of natural gradients. We evaluate the algorithm on real-world datasets containing up to 11 million data points and demonstrate that it is up to three orders of magnitude faster than the state-of-the-art while being competitive in terms of prediction performance.

  • efficient approaches to gaussian Process Classification
    Neural Information Processing Systems, 1999
    Co-Authors: Lehel Csato, Ernest Fokoue, Manfred Opper, Bernhard Schottky, Ole Winther
    Abstract:

    We present three simple approximations for the calculation of the posterior mean in Gaussian Process Classification. The first two methods are related to mean field ideas known in Statistical Physics. The third approach is based on Bayesian online approach which was motivated by recent results in the Statistical Mechanics of Neural Networks. We present simulation results showing: 1. that the mean field Bayesian evidence may be used for hyperparameter tuning and 2. that the online approach may achieve a low training error fast.

Carl Edward Rasmussen - One of the best experts on this subject based on the ideXlab platform.

  • approximations for binary gaussian Process Classification
    Journal of Machine Learning Research, 2008
    Co-Authors: Hannes Nickisch, Carl Edward Rasmussen
    Abstract:

    We provide a comprehensive overview of many recent algorithms for approximate inference in Gaussian Process models for probabilistic binary Classification. The relationships between several approaches are elucidated theoretically, and the properties of the different algorithms are corroborated by experimental results. We examine both 1) the quality of the predictive distributions and 2) the suitability of the different marginal likelihood approximations for model selection (selecting hyperparameters) and compare to a gold standard based on MCMC. Interestingly, some methods produce good predictive distributions although their marginal likelihood approximations are poor. Strong conclusions are drawn about the methods: The Expectation Propagation algorithm is almost always the method of choice unless the computational budget is very tight. We also extend existing methods in various ways, and provide unifying code implementing all approaches.

  • assessing approximations for gaussian Process Classification
    Neural Information Processing Systems, 2005
    Co-Authors: Malte Kuss, Carl Edward Rasmussen
    Abstract:

    Gaussian Processes are attractive models for probabilistic Classification but unfortunately exact inference is analytically intractable. We compare Laplace's method and Expectation Propagation (EP) focusing on marginal likelihood estimates and predictive performance. We explain theoretically and corroborate empirically that EP is superior to Laplace. We also compare to a sophisticated MCMC scheme and show that EP is surprisingly accurate.

  • assessing approximate inference for binary gaussian Process Classification
    Journal of Machine Learning Research, 2005
    Co-Authors: Malte Kuss, Carl Edward Rasmussen
    Abstract:

    Gaussian Process priors can be used to define flexible, probabilistic Classification models. Unfortunately exact Bayesian inference is analytically intractable and various approximation techniques have been proposed. In this work we review and compare Laplace's method and Expectation Propagation for approximate Bayesian inference in the binary Gaussian Process Classification model. We present a comprehensive comparison of the approximations, their predictive performance and marginal likelihood estimates to results obtained by MCMC sampling. We explain theoretically and corroborate empirically the advantages of Expectation Propagation compared to Laplace's method.

Mt Mascia - One of the best experts on this subject based on the ideXlab platform.

  • ab0656 cryoglobulin evaluation analysis of intra laboratory and inter laboratory variability
    Annals of the Rheumatic Diseases, 2018
    Co-Authors: Daniele Campioli, P Natali, D Debbia, Amelia Spinella, Gilda Sandri, Caterina Cerami, Laura Maria Scichilone, Francesco Fontana, Mt Mascia
    Abstract:

    Background Cryoglobulins (CRG) are immunoglobulins that precipitate in serum at temperatures below 37°C and resolubilize upon warming. The main reasons of interest of a clinical pathologist in the study of cryoglobulinemia are: 1) lack of standardisation in the preanalytical, analytical and postanalytical phases of the Process (Classification and reporting); 2) peculiarities of physiopathological mechanism 3) important clinical consequences. Vermeersch et al. studied these issues in 2008. To assess current practice in the detection, analysis, and reporting of cryoglobulins, a questionnaire was sent to 140 laboratories. They showed that only 36% of laboratories used standard procedures of analysis. Consequently, they concluded that standardisation was needed for cryoglobulin detection to avoid missed diagnoses and improve the comparability of results. Sargur et al. in 2010 reviewed the Classification and clinical features of cryoglobulins and suggested “best practice” guidelines for laboratory detection and identification of cryoglobulins. They particularly highlighted the relevance of preanalytical and analytical phases: maintenance of the sample at a stable temperature of 37°C, especially throughout the initial steps (collection and transportation); centrifugation and separation methods; cryoprecipitate quantification; cryoprecipitate washing techniques; immunocharacterization of cryoprecipitates especially through immunofixation techniques (considered the “gold standard”). Objectives To verify and assess the variability of laboratory Processes of CRG. Methods We checked laboratory databases of Hospital and University (Lab A and B) of Modena with long tradition in the cryoglobulin analysis (more than 6000 tests from 2002 to 2017). Concerning CRG testing, 734 patient samples were studied in both laboratories. We compared our results according to Brouet Classification into subgroups: type I, II and III. Therefore, we evaluated intra-laboratory variability, compared to previous or more frequent results. Finally, we studied inter-laboratory variability based on non-concordant laboratory reports. Results In the following table, we have represented the comparison between labs about the same patient cohort in 734 patient samples: Conclusions No data about variability in CRG analysis are reported in literature. National and international guidelines are not explicative enough. Furthermore, many doubts about Classifications are established. Our experience is unique but limited in two laboratories. Given the variability of testing conditions used in different laboratories and the lack of test standards and reference values, we confirm the need of further investigations into standardisation of CRG testing. New guidelines are fundamental, in order to optimise all phases of CRG research (pre and post analysis) and to ensure correct diagnosis and adequate treatments of the associated diseases. Disclosure of Interest None declared

  • Cryoglobulin evaluation: analysis of intra-laboratory and inter-laboratory variability
    'BMJ', 2018
    Co-Authors: Campioli Daniele, Natali P, Debbia D, Spinella A, Sandri G, Cerami C, Scichilone L, Fontana F, Mt Mascia
    Abstract:

    Background Cryoglobulins (CRG) are immunoglobulins that precipitate in serum at temperatures below 37\ub0C and resolubilize upon warming. The main reasons of interest of a clinical pathologist in the study of cryoglobulinemia are: 1) lack of standardisation in the preanalytical, analytical and postanalytical phases of the Process (Classification and reporting); 2) peculiarities of physiopathological mechanism 3) important clinical consequences. Vermeersch et al. studied these issues in 2008. To assess current practice in the detection, analysis, and reporting of cryoglobulins, a questionnaire was sent to 140 laboratories. They showed that only 36% of laboratories used standard procedures of analysis. Consequently, they concluded that standardisation was needed for cryoglobulin detection to avoid missed diagnoses and improve the comparability of results. Sargur et al. in 2010 reviewed the Classification and clinical features of cryoglobulins and suggested \u201cbest practice\u201d guidelines for laboratory detection and identification of cryoglobulins. They particularly highlighted the relevance of preanalytical and analytical phases: maintenance of the sample at a stable temperature of 37\ub0C, especially throughout the initial steps (collection and transportation); centrifugation and separation methods; cryoprecipitate quantification; cryoprecipitate washing techniques; immunocharacterization of cryoprecipitates especially through immunofixation techniques (considered the \u201cgold standard\u201d). Objectives To verify and assess the variability of laboratory Processes of CRG. Methods We checked laboratory databases of Hospital and University (Lab A and B) of Modena with long tradition in the cryoglobulin analysis (more than 6000 tests from 2002 to 2017). Concerning CRG testing, 734 patient samples were studied in both laboratories. We compared our results according to Brouet Classification into subgroups: type I, II and III. Therefore, we evaluated intra-laboratory variability, compared to previous or more frequent results. Finally, we studied inter-laboratory variability based on non-concordant laboratory reports. Results In the following table, we have represented the comparison between labs about the same patient cohort in 734 patient samples:Conclusions No data about variability in CRG analysis are reported in literature. National and international guidelines are not explicative enough. Furthermore, many doubts about Classifications are established. Our experience is unique but limited in two laboratories. Given the variability of testing conditions used in different laboratories and the lack of test standards and reference values, we confirm the need of further investigations into standardisation of CRG testing. New guidelines are fundamental, in order to optimise all phases of CRG research (pre and post analysis) and to ensure correct diagnosis and adequate treatments of the associated diseases