Xi Chen

As other reviewers have said, he's a pretty nice and chill guy, and very lenient about grading. I thought his lecturing was also pretty clear and easy to follow. He was also very receptive to questions and willing to help out. A lot of the content is mainly introductory stuff that you learn in DSA, but maybe with a bit more rigor/math behind it. Most of the topics started out really basic, and maybe got a bit complicated towards the end, but were very manageable overall. We were intending on covering NP completeness in the last lecture, but we ran out of time, so that was unfortunate. Lectures were recorded, and for the most part, they were pretty good. He did have to manually change the camera to point to the current board, which was unfortunate since sometimes he forgot. This results in him explaining parts and referring to different parts of the board that aren't visible, but usually you could infer what he was talking about The lectures followed along with the textbook so it was easy to read ahead or review things after class. Also, his lecture notes/slides were pretty comprehensive and 1:1. Sometimes, I would go through both sources and find a better explanation in the other. Homework was assigned every other week and was pretty reasonable in terms of difficulty and time commitment. Grading was fair, average was maybe ~45/50? All the problems were from the book CLRS, and…

Xi Chen is a very kind and approachable professor, which makes it difficult to be so critical. Unfortunately, he exemplifies a common failure in universities like Columbia: assuming that high intelligence translates into effective teaching. This was my second course with him, and in both cases I found his teaching to be consistently less effective than learning the material from the book(s) rather than attending lecture. Despite the class being 10–20 students, Professor Chen is largely disconnected from the students in the room and teaches as if he is delivering a pre-recorded lecture to a massive audience. Don't bother showing up to class. Watch the recordings, pause when you get stuck, and follow along in the text. One especially frustrating thing is the lack of a consistent set of course materials. For a subject Prof. Chen has taught for at least 10 years, it is inexcusable that there is no single, cohesive set of lecture notes or a clearly designated primary textbook that he sticks to. Instead, the course jumps between sources, which adds unnecessary difficulty, especially when the primary mode of learning is independent study due to the limited effectiveness of lectures. The course could also benefit from lecture slides rather then Xi Chen writing everything down on a chalkboard. Slides would not only aid Prof. Chen in teaching more effectively while structuring lectures…

This class is probably GE for most cs majors and my hopes going in is just to finish off with an A and be done with it. The course itself is very proof-based and theoretical. Prof. Chen did a excellent job breaking down the material and the homework were well-designed to prepare us for exams. Exams are fair and 80% went over in class, with 20% requiring some extra thinking that is doable if you understood the proofs in class. He is also a very approachable and understanding person in general. I strongly recommend him if you have similar expectation of the course as me. Very generous curve applied at the end.

Xi is alright, although his lectures are anything but stimulating/helpful. You can solve all of the problems using the equations in the book. You need to know some weird geometries to set up some of the problems he makes up.

His accent isnt that bad, you won't have trouble understanding him. Lectures move at a decent pace but usually there is no problem keeping up unless the problem is keeping your head up. It's easier to stay awake for the class because its only 50min (though you sacrifice your Friday mornings). His lectures usually guide you through examples, most of which are useful for the homeworks. Some of the theory-ish notes may seem trivial, but these are the things he will test on the 'multiple-choice' portion of the midterms. Midterms are pretty tricky, usually involving a rigorous knowledge of the material. There will be manipulation necessary, so you can't expect to do well just knowing equations or regurgitating homework problems. With a 1 page cheat sheet, the average on the first midterm was under 50%. The final was simple, provided that you knew the answers. 16 questions were given to us in advance and 4 were on the final. FIGURE OUT THE ANSWERS TO ALL OF THEM. He will pick the tricky ones. But make sure you know how to do them all. Homeworks sometimes were really easy, sometimes really hard. There were about 5-8 questions a week, but a few at least were simple copying from a book example or from the notes. No drops, but he is fair, not assigning homeworks during midterm weeks.

An excellent professor who taught an engaging class even during the COVID semester. He gave us biweekly psets which were very fun to solve. He was also very capable and was able to simplify complicated proofs and present them in a straightforward manner without sacrificing rigor.

Computer Science Theory is a difficult, boring subject that you will probably never need to know about. This class takes that difficulty and boringness to new heights. This is a required class for a lot of people, and it's gonna suck for a lot of people, but I DO NOT RECCOMEND TAKING THIS CLASS WITH PROFESSOR CHEN. He seems nice enough, but his teaching style and class structure is a mess. Try a different professor. Classes consist of the professor writing notes on a chalkboard and trying to explain concepts. The notes are borderline unreadable if you don't sit in the first few rows. I often found myself having to take a picture of the board with my phone and zoom in just to read it. The recordings frequently had problems such as missing audio, a completely zoomed out camera, or bizarre camera angles that make it impossible to read the board. Oftentimes if you missed the notes in class, tough luck, there's no way to look back at them. The professor has an accent that may take some getting used to, but the main issue is just the confusing and dry way he explains topics. Despite attending every lecture, I learned a lot more from YouTube than class. I recommend the channel "Easy Theory". The organization of topics is reasonable, but the organization of the class is a hot mess. Along with the problematic recordings, questions on Piazza (our Q&A board) were frequently left unanswer…

First of all, Xi still has a pretty thick accent. You can get used to this. His class is tough though. Homeworks range from not too bad to very difficult. Look to spend a lot of time on them, preferably in a group (this is encouraged by the professor). His lecture style doesn;t help much. He moves VERY quickly, but will always slow doown to answer questions. The book sucks and doesn't help much at all. You absolutely have to go to class and pay attention to even think about doing better than a B in this class. I would recommend writing down all the examples, since the book doesn't begin to tell you how to solve the homework problems. Another pain in the ass is that the class met Mon Wed Fri, at 10am. Learn the material backwards and forwards. The midterms are awful. No numbers and no BS-ing your way through them. Knowing the formulas is nowhere near enough. They are the types of problems where only symbols are given and a number should magically pop out at the end as the answer. The final is weird. 16 problems given in advance. 4 are chosen to be on the exam. Basically by the exam time you had better have solved them all or found someone who has or else you are screwed. You also have to know why your answer works. Another hint: he chooses the trickiest problems to put on the exam. Good luck, you're gonna need it.

His tests are really annoying. In addition to the problems found in all engineering tests, he adds multiple choice bullshit that cost me my grade. Class is like any engineering. You can teach yourself, but go if you feel it helps you. I didn't like the book much. Oh well, it was good enough to learn what i needed, it just took longer than it could have. The final was the only adequate test in the class. He gave you 15 questions a few weeks before, and took 4 questions from those 15 and that was the final.

Done with final, feel I have the right to write the review. First, let me say: Professor Xi Chen is so CUTE! In terms of the class, as people reviewed before me, he is rigorous about proof. He has pre-made slides, which could have some typos, but his lectures can be a little different. He is a good lecturer and very helpful during OH. Algo I covers a lot of topics in just one semester. If you think it's an upper level data structures class, then I want to tell you it's much more than that. It's like C++98 to C++17. I just want to say that we really learn a lot of different algorithms and the proofs about their correctness. The problem sets are mostly exercises from CLRS, our textbook. Some problems on them are doable, some are not (not for me, sad). I had to go to OHs to figure out where to start for some of the homework problems. Midterm and final are about the same difficulty but midterm was given shorter time. I'm not sure about grades, but the amount I learned is enough for me to not care. Although there are much more master students and a few PhDs, I think the number of undergrads taking this class is about 20. Very happy :) Some advice: read the textbook; it takes iterations (for me) to understand things. go to OH if you can; TAs are totally amazing, they see a problem and they solve it already. start the problem sets early because they could literally take 20+ hrs to do…