Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
cs-401r:assignment-5.75 [2014/10/24 16:12]
ringger [Question #5]
cs-401r:assignment-5.75 [2014/10/24 16:14]
ringger [Question #5]
Line 87: Line 87:
 * You will compute the maximum likelihood estimates for parameters $\lambda, \beta_1, \beta_2$ using the Expectation Maximization algorithm. * You will compute the maximum likelihood estimates for parameters $\lambda, \beta_1, \beta_2$ using the Expectation Maximization algorithm.
 * Construct a table formatted as shown here to capture your results. * Construct a table formatted as shown here to capture your results.
 +* Show your work separately. ​ If you show the careful work for E-step #1 and M-step #1 and you are confident about how to proceed with the next iteration, then it would be reasonable to be more terse in showing work for the subsequent steps.
  
 ^ Iteration ^ $\lambda$ ^ $\beta_1$ ^ $\beta_2$ ^ $P(Y=H | \underline{e}_1,​ \Theta)$ ^ $P(Y=H | \underline{e}_2,​ \Theta)$ ^ $P(Y=H | \underline{e}_3,​ \Theta)$ ^ $P(Y=H | \underline{e}_4,​ \Theta)$ ^ $P(Y=H | \underline{e}_5,​ \Theta)$ ^ ^ Iteration ^ $\lambda$ ^ $\beta_1$ ^ $\beta_2$ ^ $P(Y=H | \underline{e}_1,​ \Theta)$ ^ $P(Y=H | \underline{e}_2,​ \Theta)$ ^ $P(Y=H | \underline{e}_3,​ \Theta)$ ^ $P(Y=H | \underline{e}_4,​ \Theta)$ ^ $P(Y=H | \underline{e}_5,​ \Theta)$ ^
Line 116: Line 117:
  
 ==== Step 2 ==== ==== Step 2 ====
 +* Fill in the rest of row 1 as follows:
 ** E-step #2: ** E-step #2:
 *** Compute the fractional (posterior) counts for each of the five instances in the partially observed data set using the parameters computed in M-step #1. *** Compute the fractional (posterior) counts for each of the five instances in the partially observed data set using the parameters computed in M-step #1.
Line 123: Line 125:
  
 ==== Step 3 ==== ==== Step 3 ====
 +* Fill in the rest of row 2 as follows:
 ** E-step #3: ** E-step #3:
 *** Compute the fractional (posterior) counts for each of the five instances in the partially observed data set using the parameters computed in M-step #2. *** Compute the fractional (posterior) counts for each of the five instances in the partially observed data set using the parameters computed in M-step #2.
Line 130: Line 133:
  
 ==== Step 4 ==== ==== Step 4 ====
 +* Fill in the rest of row 3 as follows:
 ** E-step #4: ** E-step #4:
 *** Compute the fractional (posterior) counts for each of the five instances in the partially observed data set using the parameters computed in M-step #3. *** Compute the fractional (posterior) counts for each of the five instances in the partially observed data set using the parameters computed in M-step #3.
-* Show your work separately. ​ If you show the careful work for E-step #1 and M-step #1 and you are confident about how to proceed with the next iteration, then it would be reasonable to be more terse in showing work for the subsequent steps. 
  
 === Question #6 === === Question #6 ===
cs-401r/assignment-5.75.txt ยท Last modified: 2014/10/24 16:14 by ringger
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0