What are the mathematics that are necessary to understand Hidden Markov Models? Matrix Algebra? Linear algebra? Calculus? Bayesian statistcs?


Basic probability theory (i.e. when to sum versus multiply probabilities) is the only essential requirement. For Bayesian HMMs you would of course need to understand Bayesian inference, but HMMs do not have to be Bayesian. People get intimidated by Hidden Markov Models but they're actually quite easy to understand if you work through a few examples.

  • 1
    $\begingroup$ I do agree. However in most textbooks, basic linear algebra will be used. If you want to be prepared to Baum Welsch, have a look on EM algorithm. $\endgroup$ – Elvis Dec 7 '12 at 22:35
  • $\begingroup$ What does "to Baum Welsch" mean? What is EM algorithm? $\endgroup$ – histelheim Dec 8 '12 at 0:28
  • $\begingroup$ @AronLindberg, Baum Welch is a standard algorithm for learning the parameters of the HMM. You can find plenty more information on it by doing a google search. $\endgroup$ – jerad Dec 9 '12 at 20:28

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.