Stochastic expectation maximization with variance reduction

Expectation-Maximization (EM) is a popular tool for learning latent variable models, but the vanilla batch EM does not scale to large data sets because the whole data set is needed at every E-step. Stochastic Expectation Maximization (sEM) reduces the cost of E-step by stochastic approximation. Howe...

Olles dieđut

Bibliográfalaš dieđut
Váldodahkkit: Chen, J, Zhu, J, Teh, Y, Zhang, T
Materiálatiipa: Conference item
Almmustuhtton: Massachusetts Institute of Technology Press 2018