Acceleration of the EM algorithm using the Vector Aitken method and its Steffensen form (Q523777): Difference between revisions
From MaRDI portal
Created a new Item |
ReferenceBot (talk | contribs) Changed an Item |
||
(5 intermediate revisions by 5 users not shown) | |||
Property / author | |||
Property / author: Xu Guo / rank | |||
Property / author | |||
Property / author: Xu Guo / rank | |||
Normal rank | |||
Property / review text | |||
The authors propose an acceleration expectation-maximization (EM) and vector Aitken (VA) accelerated EM algorithms, whose convergence speed is faster than that of the EM algorithm. The algorithms are based on the use of the VA method. The VA-accelerated EM algorithm does not use the information matrix but only uses the sequence of estimates obtained from iterations of the EM algorithm, thus it keeps the flexibility and simplicity of the EM algorithm. Th. Steffensen's iterative process is also considered and the Steffensen form of the VA-accelerated EM algorithm is provided. It is shown that the reform process has a quadratic convergence. Some numerical examples are presented to illustrate the efficiency of the proposed methods. | |||
Property / review text: The authors propose an acceleration expectation-maximization (EM) and vector Aitken (VA) accelerated EM algorithms, whose convergence speed is faster than that of the EM algorithm. The algorithms are based on the use of the VA method. The VA-accelerated EM algorithm does not use the information matrix but only uses the sequence of estimates obtained from iterations of the EM algorithm, thus it keeps the flexibility and simplicity of the EM algorithm. Th. Steffensen's iterative process is also considered and the Steffensen form of the VA-accelerated EM algorithm is provided. It is shown that the reform process has a quadratic convergence. Some numerical examples are presented to illustrate the efficiency of the proposed methods. / rank | |||
Normal rank | |||
Property / reviewed by | |||
Property / reviewed by: Abdallah Bradji / rank | |||
Normal rank | |||
Property / Mathematics Subject Classification ID | |||
Property / Mathematics Subject Classification ID: 65B05 / rank | |||
Normal rank | |||
Property / zbMATH DE Number | |||
Property / zbMATH DE Number: 6707213 / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
expectation-maximization algorithm | |||
Property / zbMATH Keywords: expectation-maximization algorithm / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
vector Aitken method | |||
Property / zbMATH Keywords: vector Aitken method / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
Steffensen iterative method | |||
Property / zbMATH Keywords: Steffensen iterative method / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
convergence rate | |||
Property / zbMATH Keywords: convergence rate / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
acceleration | |||
Property / zbMATH Keywords: acceleration / rank | |||
Normal rank | |||
Property / zbMATH Keywords | |||
numerical examples | |||
Property / zbMATH Keywords: numerical examples / rank | |||
Normal rank | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Property / full work available at URL | |||
Property / full work available at URL: https://doi.org/10.1007/s10255-017-0648-3 / rank | |||
Normal rank | |||
Property / OpenAlex ID | |||
Property / OpenAlex ID: W2614497824 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Extrapolation methods theory and practice / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4139463 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Sampling-Based Approaches to Calculating Marginal Densities / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Conjugate Gradient Acceleration of the EM Algorithm / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4363981 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Acceleration of the EM and ECM algorithms using the Aitken \(\delta^2\) method for log-linear models with partially classified data / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4328418 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4844338 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: The ECME algorithm: A simple extension of EM and ECM with faster monotone convergence / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Parameter expansion to accelerate EM: the PX-EM algorithm / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q3949804 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Maximum likelihood estimation via the ECM algorithm: A general framework / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q4246485 / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Acceleration of the EM algorithm using the vector epsilon algorithm / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: On the convergence properties of the EM algorithm / rank | |||
Normal rank | |||
links / mardi / name | links / mardi / name | ||
Latest revision as of 17:54, 13 July 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Acceleration of the EM algorithm using the Vector Aitken method and its Steffensen form |
scientific article |
Statements
Acceleration of the EM algorithm using the Vector Aitken method and its Steffensen form (English)
0 references
21 April 2017
0 references
The authors propose an acceleration expectation-maximization (EM) and vector Aitken (VA) accelerated EM algorithms, whose convergence speed is faster than that of the EM algorithm. The algorithms are based on the use of the VA method. The VA-accelerated EM algorithm does not use the information matrix but only uses the sequence of estimates obtained from iterations of the EM algorithm, thus it keeps the flexibility and simplicity of the EM algorithm. Th. Steffensen's iterative process is also considered and the Steffensen form of the VA-accelerated EM algorithm is provided. It is shown that the reform process has a quadratic convergence. Some numerical examples are presented to illustrate the efficiency of the proposed methods.
0 references
expectation-maximization algorithm
0 references
vector Aitken method
0 references
Steffensen iterative method
0 references
convergence rate
0 references
acceleration
0 references
numerical examples
0 references