Daily Dozen:
1. What is a Markov chain?
2. What is a first-order Markov chain? What is a higher-order Markov chain?
3. If you know the parameters of your model, and you want to calculate the probability of your sequence, which algorithm do you use?
4. If you have an observed output sequence and you want to know the underlying state sequence, which algorithm do you use?
5. Why are all transitions not possible in higher-order Markov chains?
6. In the dog/weather example, what is hidden? What is emitted?
7. How could HMMs be used in speech processing? Why are they so good with string analysis, especially segmentation?
8. What is proof by induction?
9. Once again, what does "memoryless" mean?
10. What is a transition matrix? What are emission probabilities?
1. What is a Markov chain?
2. What is a first-order Markov chain? What is a higher-order Markov chain?
3. If you know the parameters of your model, and you want to calculate the probability of your sequence, which algorithm do you use?
4. If you have an observed output sequence and you want to know the underlying state sequence, which algorithm do you use?
5. Why are all transitions not possible in higher-order Markov chains?
6. In the dog/weather example, what is hidden? What is emitted?
7. How could HMMs be used in speech processing? Why are they so good with string analysis, especially segmentation?
8. What is proof by induction?
9. Once again, what does "memoryless" mean?
10. What is a transition matrix? What are emission probabilities?