I am reading Variational Inference: A Review for Statisticians by Blei, Kucukelbir and McAuliffe. I am having a hard time following some of the steps in Section 3.2

More precisely, they state that $\phi_{ik}$ = $E$$[c_{ik};\phi_{i}$]. I am not sure I follow this. I understand that $c_{i}$ is an indicator vector and that $q(c_{i};\phi_{i})$ is the distribution on the mixture assignment of the $i$th observation vector. The statement in that particular statement goes onto say that because $c_{i}$ is an indicator vector, this follows. I am not sure I follow it.

Would appreciate an explanation of this.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.