This preview shows page 1. Sign up to view the full content.
Unformatted text preview: DGMs – Lecture 4  Jan 23rd, 2013 page 486 (of 239) HMMs HMMs as GMs Other HMM queries What HMMs can do MPE Summ Mealy vs. Moore Machines
Mealy vs. Moore ﬁnite state automata:
Input
Symbol
State Logic Input
Symbol Output
Symbol State Memory Output
Symbol Logic Logic
Memory Moore machine has only one possible output for each state — the
output is a function only of the current state.
Mealy machine has only one possible output for each stateinput
pair — given current state, the input determines both the next state
and the current output.
Mealy machine: given a current state, and a next state (determined
by input), there is an output. Prof. Jeﬀ Bilmes EE596A/Winter 2013/DGMs – Lecture 4  Jan 23rd, 2013 page 486 (of 239) HMMs HMMs as GMs Other HMM queries What HMMs can do MPE Summ Mealy vs. Moore Machines
Mealy vs. Moore ﬁnite state automata:
Input
Symbol
State Logic Input
Symbol Output
Symbol State Memory Output
Symbol Logic Logic
Memory Moore machine has only one possible output for each state — the
output is a function only of the current state.
Mealy machine has only one possible output for each stateinput
pair — given current state, the input determines both the next state
and the current output.
Mealy machine: given a current state, and a next state (determined
by input), there is an output.
Hence, we can think of a Mealy machine as one where the output is
a function of the transition between two states.
Prof. Jeﬀ Bilmes EE596A/Winter 2013/DGMs – Lecture 4  Jan 23rd, 2013 page 486 (of 239) HMMs HMMs as GMs Other HMM queries What HMMs can do MPE Summ Mealy vs. Moore HMMs
Mealy (or Jelinek) HMMs are
ones where the observation
distribution is a function of the
transition between states. Prof. Jeﬀ Bilmes EE596A/Winter 2013/DGMs – Lecture 4  Jan 23rd, 2013 page 487 (of 239) HMMs HMMs as GMs Other HMM queries What HMMs can do MPE Summ Mealy vs. Moore HMMs
Mealy (or Jelinek) HMMs are
ones where the observation
distribution is a function of the
transition between states.
Markov transition diagram
often shown with symbols on
the edges between state nodes
(note, this is not a graphical
model, rather a labeled
statetransition diagram). Prof. Jeﬀ Bilmes 1 2 all but
“(” or “)” all but
“(” or “)” “(” “)” EE596A/Winter 2013/DGMs – Lecture 4  Jan 23rd, 2013 page 487 (of 239) HMMs HMMs as GMs Other HMM queries What HMMs can do MPE Summ Mealy vs. Moore HMMs
Mealy (or Jelinek) HMMs are
ones where the observation
distribution is a function of the
transition between states. Moore (or Rabiner) HMMs are
ones where the observation is a
function of the state itself. Markov transition diagram
often shown with symbols on
the edges between state nodes
(note, this is not a graphical
model, rather a labeled
statetransition diagram). Prof. Jeﬀ Bilmes 1 2 all but
“(” or “)” all but
“(” or “)” “(” “)” EE596A/Winter 2013/DGMs – Lecture 4  Jan 23rd, 2013 page 487 (of 239) HMMs HMMs as GMs Other HMM queries What HMMs can do MPE Summ Mealy vs. Moore HMMs Markov transition diagram
often shown with symbols on...
View
Full
Document
This document was uploaded on 04/05/2014.
 Winter '14

Click to edit the document details