1. Hidden Markov Model LR Rabiner
(Thu)
Computational Models of Intelligence
Joon Shik Kim

2. Discrete Markov Process

3. Variables S: state (rain, cloudy, sunny) O: observation (umbrella)
How to infer the weather sequences based on only the observations?

4. Three Basic Problems for HMM
Problem 1: Given the observation sequence O and model ,
how do we efficiently compute ,
the probability of the observation sequence, given the model?
Problem 2: Given the observation sequence O and the model , how do we choose a corresponding state sequence Q which is optimal in some meaningful sense (i.e., best “explains” the observations)?

5. Three Basic Problems for HMM
Problem 3: How do we adjust the model parameters to maximize ?

6. Solution to Problem 1
Forward-Backward Procedure

7. Solution to Problem 2
Viterbi algorithm

8. Solution to Problem 3 Baum-Welch method
EM (expectation-modification) method