I have not given a homework assignment this week. This will give
people who are new to covariance and such concepts time to go over the
notes on the web page. I recently expanded and hopefully clarified
the notes quite a bit. (I am sure there are mistakes -- I'll be
grateful for any corrections.)
Next week, I will give the real-valued boosting algorithm and
analysis, and talk about learning "Generalized Additive Models" from
statistics.
I also plan to give a harder-than-usual homework assignment next week.
However, as an alternative to the theory problems, I will also give a
related programming assignment that people can do instead, if they
prefer.
- Adam