Synonyms
Definition
Expectation propagation is an algorithm for Bayesian machine learning (see Bayesian Methods). It tunes the parameters of a simpler approximate distribution (e.g., a Gaussian) to match the exact posterior distribution of the model parameters given the data. Expectation propagation operates by propagating messages, similar to the messages in (loopy) belief propagation (see Graphical Models). Whereas messages in belief propagation correspond to exact belief states, messages in expectation propagation correspond to approximations of the belief states in terms of expectations, such as means and variances. It is a deterministic method especially well-suited to large databases and dynamic systems, where exact methods for Bayesian inference fail and Monte Carlo methods are far too slow.
Motivation and Background
One of the main problems for Bayesian methodsare their computational expense: computation of the exact posterior, given the observed data, typically requires...
Recommended Reading
Csató, L. (2002). Gaussian processes – iterative sparse approximations. PhD thesis, Aston University, Birmingham, UK.
Herbrich, R., & Graepel, T. (2006). TrueSkill: A Bayesian skill rating system. (Tech. Rep. No. MSR-TR-2006-80). Cambridge, UK: Microsoft Research.
Heskes, T., Opper, M., Wiegerinck, W., Winther, O., & Zoeter, O. (2005). Approximate inference with expectation constraints. Journal of Statistical Mechanics: Theory and Experiment, 11, P11015-1–P11015-24.
Heskes, T., & Zoeter, O. (2002). Expectation propagation for approximate inference in dynamic Bayesian networks. In A. Darwiche & N. Friedman (Eds.), Proceedings of the 18th conference on uncertainty in artificial intelligence (pp. 216–223). San Francisco: Morgan Kaufmann.
Minka, T. (2001). A family of algorithms for approximate Bayesian inference. PhD thesis, Cambridge, MA: MIT.
Minka, T. (2005). Divergence measures and message passing. (Tech. Rep. NO. MSR-TR-2005-173), Cambridge, UK: Microsoft Research.
Minka, T., & Lafferty, J. (2002). Expectation-propogation for the generative aspect model. In A. Darwiche & N. Friedman (Eds.), Proceedings of the 18th conference on uncertainty in artificial intelligence (pp. 352–359). San Francisco: Morgan Kaufmann.
Opper, M., & Winther, O. (2001). Tractable approximations for probabilistic models: The adaptive Thouless-Anderson-Palmer mean field approach. Physical Review Letters, 86, 3695–3699.
Seeger, M. (2005). Expectation propagation for exponential families (Tech. Rep.). Berkeley, CA: University of California.
Welling, M., Minka, T., & Teh, Y. (2005). Structured region graphs: Morphing EP into GBP. In F. Bacchus & T. Jaakkola (Eds.), Proceedings of the 21st conference on uncertainty in artificial intelligence (UAI) (pp. 609–614). Arlington, VA: AUAI Press.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this entry
Cite this entry
Heskes, T. (2011). Expectation Propagation. In: Sammut, C., Webb, G.I. (eds) Encyclopedia of Machine Learning. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-30164-8_290
Download citation
DOI: https://doi.org/10.1007/978-0-387-30164-8_290
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-30768-8
Online ISBN: 978-0-387-30164-8
eBook Packages: Computer ScienceReference Module Computer Science and Engineering