Name: __________________________________________________

Checking In

How well are you attending to your own wellness this week?

      Not at all ☐— — — — — — — — — —☐ Very well

How much time did you spend on CS159 outside of class this week? ____________________________________

Shared Reading

The following key terms were important to this week’s reading:

language model                          training set
n-gram                                  development set
bigram                                  test set
trigram                                 perplexity
chain rule                              sparsity
Markov assumption                       zeros
maximum likelihood estimation           closed/open vocabulary
normalize                               OOV word
relative frequency                      Laplace smoothing
extrinsic evaluation                    discounting
intrinsic evaluation                    backoff
                                        interpolation

1. Put a “*” next to the terms you feel most comfortable with, and a “?” next to the terms you still have questions about.

2. Why are probabilities usually computed in log space?
   
   
   
   
   
   

3. Explain why NLP experiments generally need a training set, a dev set, and a test set.
   
   
   
   
   
   

4. Explain the difference between backoff and interpolation.
   
   
   
   
   
   

Optional Reading

Which reading topic did you choose?

Put notes here on your group’s discussion

Reflection

How prepared were you for today’s class?

      Not at all ☐— — — — — — — — — —☐ Very well

What would you like me to know about how your groups’ discussions went?