Computer Science · Columbia University
6 problem sets graded on completion for 20%, two midterms for 30% each, one take home coding exam for 20%.
As long as you show that you put in effort, you can get a good grade on assignments. However, I think this professor is actually one of the hardest to get a high overall grade from, more than you'd expect.
Spent 3-5 hours per problem set, ~8 hours studying for the midterm, 2.5 hours/week in lectures, and about 20 hours on the final project.
Very fair workload that helped me structure my independent research better so I wasn't overwhelmed at the end of the semester.
Tbh, the material is difficult, but it's the nature of it. I went to his office hours only sometimes and he's always always always helpful! It wasn't too difficult that I had to be at his OH every single time. Everything is definitely doable on your own! The grades are mostly if you finished the coding assignment, no hidden testing or anything, very generous grades given. Sometimes I felt like I didn't even deserve the grades he gave me, but he sees your effort and that's what really matter!! Any of his class is highly recommended!!!!!!!!
6 bi-weekly psets graded on completion; 2 in-class midterms (one being a non-cumulative final) and 1 take-home coding assessment.
The professor has a kind demeanor and teaches well, but I am dissatisfied with the way the class is managed. By accommodating repeated requests for make-up exams, it seems that the rest of the students are put at a disadvantage — especially when make-up exams are held up to ten days later, giving some students a longer preparation period. It feels like too much consideration is being given to convenience, which leads to issues of fairness. Additionally, I don’t quite understand why recovery exams are allowed when everyone originally took the same test. I believe this creates a reverse discrimination against those who performed well the first time.
Brian is enthusiastic, kind, understanding, and receptive to feedback. I took his class because of the CULPA reviews, not knowing anything about the material. It turns out, the course content is the most useful out of all the CS courses I've taken so far. If you want to understand NVIDIA, GPU programming (essential in AI), and advanced robot algorithms, take this course! The lectures were recorded, though most people showed up to class because he's an engaging lecturer. There were 3 problem sets, each consisting of a conceptual and coding portion which were very doable by using the lecture slides and going to OH. For each problem set, he sets aside one class time for the class to work on it together and ask him questions. The midterm exam was harder than the problem sets but definitely doable. The final research project felt a bit high-pressure because we only had a month to do the whole thing, but it was a useful experience to get a glimpse of what robotics/GPU research is like. You should take this course if you don't want to be worried about your grade (he's generous) and want to learn in a super collaborative, exciting environment.
Professor Plancher is a great lecturer and brought really interesting guest speakers who provided a very well rounded introduction to various aspects of computer science development. He is really accommodating, kind and has his students' best interest at heart. He is passionate about his teaching which is evident in class.
I didn't expect much going into this class, I just wanted to dip my toes into the water of robotics. However, thanks to Brian, I am deep in the ocean! Brian is very understanding and he cares about your education! He makes the materials that seem very complicated to very simple for you to digest even just a little bit. He's honestly a great mentor and a kind human being that will make you feel cared for and continue to push through the obstacles. This course is pretty similar to AI but less of the math, more of the practice of the search algorithm! I was just looking back at my notes and his slides the other day for another project and admire how much I've learnt from this class.