Expectation-Maximization

Greatest expectationsIt is an algorithm for finding maximum likelihood estimates and maximum a posteriori estimates of parameters in a probability model based on unobservable dependent variables.

The maximum expectation algorithm is often used in the field of data clustering in machine learning and computer vision. It is calculated alternately in two steps:

• Calculate the expected value E: Use the existing estimates of the hidden variables to calculate the maximum likelihood estimate;
• Maximize M: Calculate the parameter value based on the maximum likelihood value obtained in the expected calculation, and use it in the next expectation calculation. The process is repeated alternately.