3
$\begingroup$

In the sources I consulted it states that in the Baum Welch algorithm the reestimate of the initial probability of state $i$ of the HMM is $\pi^*_i = \gamma_i(1)$. But $\gamma_i(t)$ is the probability of being in state ${\displaystyle i}$ at time ${\displaystyle t}$ given the observed sequence ${\displaystyle Y}$ and the parameters ${\displaystyle \theta }$ (quote wiki)

So, then why does this probability not need to be normalised like so? :

$$\pi^*_i = \frac{\gamma_i(1)}{\sum_{j = 1}^N \gamma_j(1)}$$

After all normalizing is what is done for the reestimate of the transition probabilities and the emission probabilities too.

$\endgroup$
4
$\begingroup$

It is defined to be a probability. A probability is by definition already normalized. In particular, we are guaranteed that

$$\sum_{j=1}^N \gamma_j(1) = 1,$$

as there are only $N$ possibilities for the state that you're in at time $1$, and these $N$ cases have no overlap.

$\endgroup$

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.