Results 231 to 240 of about 684,507 (282)
Some of the next articles are maybe not open access.
Biometrics, 1993
Summary: A modification of the EM algorithm is proposed for situations in which the maximization of the ``complete data'' likelihood function does not have a closed-form solution. Self-consistency of the modified algorithm is established. Application to carcinogenicity experiments is illustrated, and the results of a simulation study comparing the ...
Rai, S. N., Matthews, D. E.
openaire +1 more source
Summary: A modification of the EM algorithm is proposed for situations in which the maximization of the ``complete data'' likelihood function does not have a closed-form solution. Self-consistency of the modified algorithm is established. Application to carcinogenicity experiments is illustrated, and the results of a simulation study comparing the ...
Rai, S. N., Matthews, D. E.
openaire +1 more source
Deterministic annealing EM algorithm
Neural Networks, 1998This paper presents a deterministic annealing EM (DAEM) algorithm for maximum likelihood estimation problems to overcome a local maxima problem associated with the conventional EM algorithm. In our approach, a new posterior parameterized by `temperature' is derived by using the principle of maximum entropy and is used for controlling the annealing ...
N, Ueda, R, Nakano
openaire +2 more sources
2001
Abstract Finding maximum likelihood estimates usually requires a numerical method. Classical techniques such as the Newton–Raphson and Gauss–Newton algorithms have been the main tools for this purpose. The motivation for such techniques generally comes from calculus.
openaire +2 more sources
Abstract Finding maximum likelihood estimates usually requires a numerical method. Classical techniques such as the Newton–Raphson and Gauss–Newton algorithms have been the main tools for this purpose. The motivation for such techniques generally comes from calculus.
openaire +2 more sources
An EM Algorithm for Capsule Regression
Neural Computation, 2021We investigate a latent variable model for multinomial classification inspired by recent capsule architectures for visual object recognition (Sabour, Frosst, & Hinton, 2017 ). Capsule architectures use vectors of hidden unit activities to encode the pose of visual objects in an image, and they use the lengths of these vectors to encode the ...
openaire +3 more sources
EM Algorithm for Segregation Analysis
Biometrical Journal, 1992AbstractWe consider the problem of estimating segregation ratios in families based on ascertainment through affected children, formulate it as an incomplete problem and work out the EM algorithm for maximum likelihood estimation of segregation ratios. We treat both the cases of known and unknown ascertainment probability. We also derive expressions for
Achuthan, N. R., Krishnan, T.
openaire +2 more sources
Acceleration of the EM algorithm: P-EM versus epsilon algorithm
Computational Statistics & Data Analysis, 2012zbMATH Open Web Interface contents unavailable due to conflicting licenses.
Berlinet, Alain F., Roland, Christophe
openaire +1 more source
Local-EM and the EMS Algorithm
Journal of Computational and Graphical Statistics, 2011The use of local likelihood methods (Tibshirani and Hastie 1987; Loader 1999) in the presence of data that are either interval or area censored leads naturally to the consideration of EM-type strategies, or rather local-EM algorithms. In this article we consider a class of local-EM algorithms suitable for density or intensity estimation in the temporal
Chun-Po Steve Fan +2 more
openaire +1 more source
Statistics and Computing, 2013
zbMATH Open Web Interface contents unavailable due to conflicting licenses.
openaire +2 more sources
zbMATH Open Web Interface contents unavailable due to conflicting licenses.
openaire +2 more sources
Proceedings of the IEEE, 1988
The estimate-maximize (EM) algorithm is an iterative method for finding maximum-likelihood parameter estimates from incomplete data. The authors develop an extension of the EM algorithm that may be useful in accelerating the algorithm and in simplifying the computations involved. The extension works with an intermediate complete data specification, and
M. Segal, E. Weinstein
openaire +1 more source
The estimate-maximize (EM) algorithm is an iterative method for finding maximum-likelihood parameter estimates from incomplete data. The authors develop an extension of the EM algorithm that may be useful in accelerating the algorithm and in simplifying the computations involved. The extension works with an intermediate complete data specification, and
M. Segal, E. Weinstein
openaire +1 more source
EM*: An EM Algorithm for Big Data
2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA), 2016Existing data mining techniques, more particularly iterative learning algorithms, become overwhelmed with big data. While parallelism is an obvious and, usually, necessary strategy, we observe that both (1) continually revisiting data and (2) visiting all data are two of the most prominent problems especially for iterative, unsupervised algorithms like
Hasan Kurban +2 more
openaire +1 more source