Zobrazeno 1 - 10
of 160
pro vyhledávání: '"D. M. Titterington"'
Publikováno v:
Computational Statistics & Data Analysis. 93:246-254
A new transdimensional Sequential Monte Carlo (SMC) algorithm called SMCVB is proposed. In an SMC approach, a weighted sample of particles is generated from a sequence of probability distributions which 'converge' to the target distribution of intere
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8b598b0cfb81171e6d5c1b615a58ae99
https://doi.org/10.1016/j.csda.2015.03.006
https://doi.org/10.1016/j.csda.2015.03.006
Publikováno v:
Philosophical Magazine. 92:50-63
New statistical approaches to hyperparameter estimation by means of the expectation-maximization (EM) algorithm and loopy belief propagation (LBP) are given for Bayesian image modeling from the standpoint of statistical-mechanical informatics. In the
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::62180ed1ac0e2c4fc67f57bffad6ca57
https://doi.org/10.1080/14786435.2011.624558
https://doi.org/10.1080/14786435.2011.624558
Publikováno v:
Journal of the American Statistical Association. 104:1597-1608
Conventional distance-based classifiers use standard Euclidean distance, and so can suffer from excessive volatility if vector components have heavy-tailed distributions. This difficulty can be alleviated by replacing the L2 distance by its L1 counte
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1b3136de35a9c2559c8addf22ec69d63
https://doi.org/10.1198/jasa.2009.tm08107
https://doi.org/10.1198/jasa.2009.tm08107
Publikováno v:
Journal of the Royal Statistical Society Series B: Statistical Methodology. 71:783-803
SummaryMany contemporary classifiers are constructed to provide good performance for very high dimensional data. However, an issue that is at least as important as good classification is determining which of the many potential variables provide key i
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7a1a77271326eea28c31a005f25992c3
https://doi.org/10.1111/j.1467-9868.2009.00701.x
https://doi.org/10.1111/j.1467-9868.2009.00701.x
Autor:
Clare A. McGrory, D. M. Titterington
Publikováno v:
Australian & New Zealand Journal of Statistics. 51:227-244
The variational approach to Bayesian inference enables simultaneous estimation of model parameters and model complexity. An interesting feature of this approach is that it also leads to an automatic choice of model complexity. Empirical results from
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2489c586ecc8da1317e483db4c3a08c2
https://doi.org/10.1111/j.1467-842x.2009.00543.x
https://doi.org/10.1111/j.1467-842x.2009.00543.x
Publikováno v:
Journal of the American Statistical Association
Journal of the American Statistical Association, 2009, 104 (485), pp.263-273. ⟨10.1198/jasa.2009.0125⟩
Journal of the American Statistical Association, Taylor & Francis, 2009, 104 (485), pp.263-273. ⟨10.1198/jasa.2009.0125⟩
Journal of the American Statistical Association, 2009, 104 (485), pp.263-273. ⟨10.1198/jasa.2009.0125⟩
Journal of the American Statistical Association, Taylor & Francis, 2009, 104 (485), pp.263-273. ⟨10.1198/jasa.2009.0125⟩
The k-nearest-neighbour procedure is a well-known deterministic method used in supervised classification. This paper proposes a reassessment of this approach as a statistical technique derived from a proper probabilistic model; in particular, we modi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8bc925f33b022419f68c5637bc948f1d
https://doi.org/10.1198/jasa.2009.0125
https://doi.org/10.1198/jasa.2009.0125
Publikováno v:
Statistics and Computing. 19:329-340
Hidden Markov random field models provide an appealing representation of images and other spatial problems. The drawback is that inference is not straightforward for these models as the normalisation constant for the likelihood is generally intractab
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::ea97f1c2d5ee381dfea0abe8faf0f166
https://doi.org/10.1007/s11222-008-9095-6
https://doi.org/10.1007/s11222-008-9095-6
Autor:
D. M. Titterington, Kazuyuki Tanaka
Publikováno v:
Journal of Physics A: Mathematical and Theoretical. 40:11285-11300
We calculate analytically a statistical average of trajectories of an approximate expectation-maximization (EM) algorithm with generalized belief propagation (GBP) and a Gaussian graphical model for the estimation of hyperparameters from observable d
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::691f7dafa2d8e5b3250af14306727c1a
https://doi.org/10.1088/1751-8113/40/37/007
https://doi.org/10.1088/1751-8113/40/37/007
Autor:
Clare A. McGrory, D. M. Titterington
Publikováno v:
Computational Statistics & Data Analysis. 51:5352-5367
Variational methods, which have become popular in the neural computing/machine learning literature, are applied to the Bayesian analysis of mixtures of Gaussian distributions. It is also shown how the deviance information criterion, (DIC), can be ext
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::388f5554bafa2ce26237cadd0b855da1
https://doi.org/10.1016/j.csda.2006.07.020
https://doi.org/10.1016/j.csda.2006.07.020
Publikováno v:
Statistics and Computing. 15:31-41
As a result of their good performance in practice and their desirable analytical properties, Gaussian process regression models are becoming increasingly of interest in statistics, engineering and other fields. However, two major problems arise when
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::37d3385b8b20f2ca422800f0dd948341
https://doi.org/10.1007/s11222-005-4787-7
https://doi.org/10.1007/s11222-005-4787-7