Expectation propagation (EP) is a technique in Bayesian machine learning.

EP finds approximations to a probability distribution. It uses an iterative approach that uses the factorization structure of the target distribution. It differs from other Bayesian approximation approaches such as variational Bayesian methods.

. Expectation propagation achieves this approximation by minimizing the Kullback–Leibler divergence. instead.

being equal to the mean and the covariance, respectively; this is called moment matching.

Applications

Expectation propagation via moment matching plays a vital role in approximation for indicator functions that appear when deriving the message passing equations for TrueSkill.

References

External links

• Minka's EP papers

• List of papers using EP