Synonyms
Related Concepts
Definition
The expectation-maximization algorithm iteratively maximizes the likelihood of a training sample with respect to unknown parameters of a probability model under the condition of missing information. The training sample is assumed to represent a set of independent realizations of a random variable defined on the underlying probability space.
Background
One of the main paradigms of statistical pattern recognition and Bayesian inference is to model the relation between the observable features \(x\in \mathcal {X}\) of an object and its hidden state \(y\in \mathcal {Y}\) by a joint probability measure p(x, y). This probability measure is, however, often known only up to some parameters θ ∈ Θ. It is thus necessary to estimate these parameters from a training sample, which is assumed to represent a sequence of independent realizations of a random variable. If, ideally,...
This is a preview of subscription content, log in via an institution.
References
Meintrup D, Schäffler S (2005) Stochastik. Springer, Berlin
Papoulis A (1990) Probability and statistics. Prentice-Hall, Englewood Cliffs
Minka T (1998) Expectation-maximization as lower bound maximization. Tutorial, MIT, Cambridge, MA
Dellaert F (2002) The expectation maximization algorithm. Technical Report GIT-GVU-02-20, Georgia Institute of Technology
Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc Ser B 39(1):1–38
Schlesinger MI (1968) The interaction of learning and self-organization in pattern recognition. Kibernetika 4(2):81–88
Sundberg R (1974) Maximum likelihood theory for incomplete data from an exponential family. Scand J Stat 1(2):49–58
McLachlan GJ, Krishnan T (1997) The EM algorithm and extensions. Wiley, New York
Schlesinger MI, Hlavac V (2002) Ten lectures on statistical and structural pattern recognition. Kluwer Academic Publishers, Dodrecht
Bishop CM (2006) Pattern recognition and machine learning. Springer, Nw York
Cheng Y (1995) Mean shift, mode seeking, and clustering. IEEE Trans Pattern Anal Mach Intell 17(8):790–799
Jelinek F (1998) Statistical methods for speech recognition. MIT, Cambridge, MA
Li SZ (2009) Markov random field modeling in image analysis. Advances in pattern recognition, 3rd edn. Springer, London
Author information
Authors and Affiliations
Corresponding author
Section Editor information
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this entry
Cite this entry
Flach, B., Hlavac, V. (2020). Expectation-Maximization Algorithm. In: Computer Vision. Springer, Cham. https://doi.org/10.1007/978-3-030-03243-2_692-1
Download citation
DOI: https://doi.org/10.1007/978-3-030-03243-2_692-1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-03243-2
Online ISBN: 978-3-030-03243-2
eBook Packages: Springer Reference Computer SciencesReference Module Computer Science and Engineering