COMS W4771

There's 3 exams worth 90% of your grade. This class is a time sink if you want to do well. Lectures: expect 3+ hours per lecture to really understand (rewatching, pausing, filling gaps, re-deriving steps). Prereq math review: if you haven’t touched linear algebra / probability / calc in years (like most people), budget ~10 hours/week just to re-learn the prereqs. The prereq expectations are kind of ridiculous because basically nobody has “fresh” mastery of all of it unless they just took those classes and actually retained them. Homework: if you want real practical skills and you actually do the programming components seriously, it can be ~20 hours per homework (of which there are 6 of). Bare minimum survival mode (to not drown / to do okay): -focus on lecture material and prioritize understanding -do the theory parts of homework -do the quizzes (completion based and worth 10% of final grade) -do practice exam problems Extra mile: -do the readings (painful, but it helps a lot for exams) -do the implementation homework problems (take away some practical ability from the course)

5 Assignments : Assignments are actually easy, programing ones might take time depending on how familiar you are with matlab 2 quizzes, Midterm:easy

6 homeworks (starts out weekly, later every 2 weeks) in Python, 2 in-class multiple choice quizzes, take-home midterm, final exam. Not the worst in the department, but definitely quite heavy. Do not take this class.

This is a high value course for any aspiring SWE in the age of AI — but it will probably make you miserable, and it could be run so much better. You will learn the theoretical underpinnings of the core “meat and potatoes” supervised ML methods (linear/logistic regression, neural nets, etc.). The problem is the teaching. The professor is a bad lecturer: he nonchalantly skips steps in highly abstract math, leaves out key details, and then effectively dumps the missing reasoning into the required readings, which are insanely dense. Also, he has a serious attitude problem — he can be dismissive and unkind when students ask questions, which makes it harder to engage and learn. The readings are basically hieroglyphics. I’m not exaggerating: it can take hours to get through a single page if you actually want to understand what’s going on. There’s also a big mismatch between “what you do” and “what gets graded.” The optional homework is the only place you get hands-on practice with ML in Python, so the class barely values practical skill. You can ignore the programming parts of the homework and not worry much about exam performance, because the exams are theory-heavy and don’t care whether you can implement well. Grading is harsh and the curve is absurd — it's literally curved to the C range, which is ridiculous considering how demanding the prerequisites and workload are. I ended up…

Machine Learning by Prof. Jebara was an excellent class. There are not many graduate classes at Columbia where you feel you have learnt your money's worth. Trust me, this is one class where you will feel challenged and inspired. Prof. Jebara is good with giving intuitions about many things, picking examples from Vision, Genomics, NLP. You many find this course "hard", or "uninteresting", if you thought ML is "cool", or the sole image of Machine Learning in your head is from ML class by Andrew Ng on Coursera. This is a very maths based course, and he mentions this in the first class, and urges people to drop if they can't handle it. I guess it is fair on his part, to announce that the maths will be slighty tough, I wouldn't say even tough, just different. This course will help you a lot, if you want to read Machine Learning papers, do research in the field, because there is some very odd notation involved and the course will teach you that. My vote is take this course, if you are ready to dive into hour long lectures of Mathematical Proofs, but if you are upto it, this will be a great course.

I took Machine Learning with Professor Pe'er in Spring 2017 - I've been tardy about writing a review, but I feel it is necessary. I am genuinely in awe at how bad my experience in this class was, in seemingly every way. To start: Professor Pe'er. It is fortunate that this man is, in fact, a professor - were I to call him "Instructor Pe'er", one may get the false impression that he is capable of instructing anyone. He is not. I have never seen a human so uniquely unsuited for pedagogy as this man. His classroom delivery is a complete trainwreck - when he's not stumbling over his words or saying "ehm" for 15 seconds, the words he does say are incomprehensible, even when explaining material that is clearly within students' grasp. Unfortunately, most of his lectures did not fall into this category, as the prerequisites for this class grossly understate the necessary statistical background. As a CS student with only an intro-level stats background, I rarely had any idea what the hell was going on in lecture. Professor Pe'er's attempt to bridge this gap was laughably insufficient: he spent one day at the beginning of the semester attempting to speed through all of statistics. It didn't play well. Here is how Pe'er explained what a convex surface was: he brought a muffin into class, and passed it around the room. Here is how he explained differences in dimensionality: he printed out…

Avoid at all cost. The curve is non-existent. Mean will get you a C/C+ as oppose to B/B+ in other classes.

5 Homeworks (5-20 hours of RESEARCH, because you ain't solving these with what you "learned" in class) 1 Midterm (Average was around 50) 1 Final

I cannot stress enough how awful this teacher truly is. He will teach you absolutely nothing. He does not explain anything thoroughly. He cannot answer any questions. His slides are a total mess. If you can avoid taking this class with this teacher, I highly recommend you do.

90% is exam, he ends up giving 100 to everyone on the quiz which consists of 10% of the grade. Mean for Exams are around 50 while first one is 70. He curved the second one because it was really bad(mean around 40). Not sure if he will curve the final raw score for grade cutoff, he previously mentioned he doesn't usually have to make any "artificial modification" to grade(i.e curve). Workload is vary based on your efforts, biweekly quizzes and not graded hw.