Viterbi: dynamic programming for HMM

Recursion:

vl(i+1) =

el(xi+1) max(vk(i)akl)

si =

Most probable path

l,k=2

states

a= table in slide 51

e= emit si in state l (Durbin p.56)

1/8*.27

Previous slide

Next slide

Back to first slide

View graphic version