Problem set 1. Supervised learning.

1. a. Linear classifier: pdf.
   b. Newton's method: code.
   c. GDA has a linear decision boundary: pdf.
   d. Proof that log maximum likelihood estimates of params... pdf.
   e. Calculation of parameters code.
   f. Plot for dataset1 pdf.
   g. Plot for dataset2 pdf.
   h. Transformation for dataset1 pdf.
2. Incomplete, Positive-Only Labels
   a, b. Prove that probability... pdf.
   c, d, e. Train logistic regression on incomplete data. code
3. Poisson Regression
   a, b, c. Poisson distribution is in the exponential family pdf.
   d. Poisson regression code.