Ratings for Jordan, M
Based on 5 ratings for this instructor.
Comment: Was a standard lecturer. Pretty clear and followed the textbook. Enthusiastic about certain portions, but seemed a little out of his element on others. Exams are not particularly difficult; they do a good job of having lots of coverage and loads of questions, so that it's a test of what you know (and how well you know it, since there's a time crunch for the midterm) rather than a test of clever tricks / how well you memorized the super obscure and pointless theorem from chapter 7 section 3 subsection 1.3. So there's that going for you. Assignments weren't hard, but there were way too many and they're wayy to long. The material itself is relatively straightforward, so some psets are going to be tedious.
Submitted May 23rd, 2015
Comment: Well, Jordan was good, but there seemed to be too much hype that he didn't quite live up to. The class material was pretty good if you like probability, the exams and homeworks were challenging and fun, but lectures were very standard lectures. Jordan pretty much just took notes from the book and explained it to us exactly. Good class for the material though.
Submitted May 22nd, 2015
Comment: Lectures weren't particularly entertaining though they are clear and follows the textbook very closely. Materials were interesting if you are interested in techniques for analyzing randomized algorithms. Homework was of medium difficulty and exams were slightly easier than the homework. The staff were somewhat helpful on Piazza. It was a nice class overall.
Submitted May 21st, 2015
Comment: This is just about the best class I've taken at Berkeley (though I've only taken undergraduate classes besides this one). Michael Jordan has very good lectures, and his notes are also quite helpful for the most part. The material is super interesting for me, as the goal is to get you to understand the underpinnings of statistical learning with graphical models so that you can do research in the field. The homeworks were all interesting, though quite challenging. They featured mostly derivations/proofs, with about one implementation problem per homework. The project (which composed the rest of the grade) could have used more description ahead of time (though it's pretty standard for grad classes) -- you work on anything involving the course material, write a report, and present at a poster session at the end of the semester.
Submitted May 22nd, 2014