professor
Xiaodong Wang

Jan 2021

He's the worst. Not approachable. He will make you feel like it's your fault if you don't understand something. He was not understanding of difficulties online. He is consistently late to office hours and takes so long to grade homework and midterms that you don't know if you understand the material until you have already taken the next assessment. You'll likely end up doing fine in the class grade-wise, but not without great pain. AVOID if possible.

Jan 2021

Class is infamous at Columbia for being difficult and hard to understand. After a steep learning curve, the material becomes easier to grasp. It can be intimidating to ask questions in class but it is wise to ask them as much as you can. The professor will explain anything you ask and moves very quickly through the material. Study the class notes thoroughly and start the psets early as some later ones are very long.

Jan 2020

Don't confuse this professor Wang with the professor Wang who apparently teaches Chinese quite well. Wang is not a particularly exciting lecturer but he does convey the material fairly effectively. The material is extremely difficult and I would definitely stay away from this class unless you need it for your major or you are particularly interested in time analysis. It is really a math class in the EE department which Wang will tell you straight out so don't take this because you're interested in Circuit Analysis. The exams and HWs are very difficult if you don't have the solution manual to the point where it becomes hard to actually learn the material well since you have to really focus on doing well in the class as opposed to learning the content. He generally has a policy that if you don't do well on the midterms he will count the final instead of them changing the grading scheme from 40% midterms (2), 20% HW, 40% final to 80% final and 20% HW, so you basically must do well on the final to succeed in the class. Overall I'd say this class is extremely difficult and should be avoided unless required, but Wang is a capable professor.

May 2018

Wang Lao Shi is amazing, she is such a wonderful teacher and a wonderful person. The class is a blast and really fun to be in. Chinese is a lot of work, but Wang Lao Shi makes it super fun. She is an amazing person and a great professor.

Sep 2015

Prof Wang isn't the most animated lecturer and the room gets hot and it's Friday and it's easy to start dozing off. None of that changes the fact that information theory is BEAUTIFUL. It's easy to get lost in the mathematics (of which there is a lot - more on that in a bit). I highly recommend taking some time after you get home and try to digest what you're learning, get an intuitive feel for what entropy and channel capacity and "information" are, and you will find your mind blown time and time again. The idea that something as "dumb" and taken for granted as zipping a file is (asymptotically) equivalent to something as profound as automatically finding all the structure in a dataset is at first unfathomable but becomes obvious as you go along and has very exciting implications. Many of the proofs and algorithms you learn about (Blahut-Arimoto, reverse water-filling, Lempel-Ziv) are extremely aesthetically appealing as well. Unfortunately some of the proofs are quite tedious. Don't let the department fool you, this is not an engineering class (despite being taught by an IEEE Fellow) but a mathematics class. The registrar is lying when it says the prereq is just "probability"; the real prerequisites should be STAT4105+4107 (probability and statistics), ELEN4815 (random signals and noise; not obligatory but good to have), and real analysis. Unfortunately the class is not focused as much on interpretation as I would have liked. Cover and Thomas is an excellent textbook and a pleasure to read. I also found David MacKay's videos from Cambridge helpful for intuition for the first few weeks but be warned, this class goes a lot deeper into information theory the Cambridge class.

Sep 2015

Prof Wang isn't the most animated lecturer and the room gets hot and it's Friday and it's easy to start dozing off. None of that changes the fact that information theory is BEAUTIFUL. It's easy to get lost in the mathematics (of which there is a lot - more on that in a bit). I highly recommend taking some time after you get home and try to digest what you're learning, get an intuitive feel for what entropy and channel capacity and "information" are, and you will find your mind blown time and time again. The idea that something as "dumb" and taken for granted as zipping a file is (asymptotically) equivalent to something as profound as automatically finding all the structure in a dataset is at first unfathomable but becomes obvious as you go along and has very exciting implications. Many of the proofs and algorithms you learn about (Blahut-Arimoto, reverse water-filling, Lempel-Ziv) are extremely aesthetically appealing as well. Unfortunately some of the proofs are quite tedious. Don't let the department fool you, this is not an engineering class (despite being taught by an IEEE Fellow) but a mathematics class. The registrar is lying when it says the prereq is just "probability"; the real prerequisites should be STAT4105+4107 (probability and statistics), ELEN4815 (random signals and noise; not obligatory but good to have), and real analysis. Unfortunately the class is not focused as much on interpretation as I would have liked. Cover and Thomas is an excellent textbook and a pleasure to read. I also found David MacKay's videos from Cambridge helpful for intuition for the first few weeks but be warned, this class goes a lot deeper into information theory the Cambridge class.

Dec 2014

With all the negative reviews I think I'll come out and say that I *appreciate* Professor Wang. I don't *like* him - I don't think anybody likes him, and to be honest he doesn't give a shit about students either. He cracked exactly one joke during the semester, completely deadpan. Honestly I would not trust Vallancourt to teach this class. Professor Wang presents the material in a clear, systematic manner, and test problems are basically the same as homeworks with the numbers changed. Some of the material is difficult but gives you a solid background for future signal processing courses. I especially appreciated that proved all the theorems he presented. But be careful, it's not for nothing - the exams have proofs and you will probably be asked to prove Parseval's theorem at some point. If you ever have a chance, look up the research he does too - he's an IEEE Fellow and gets millions of dollars of NSF funding.

Dec 2011

Unless your engineering major requires it, try to avoid taking Signals and Systems with Professor Wang. First off, this man has no soul. According to some of my classmates, he merely acted like a "typical Chinese professor," but I have had professors from the Mainland/Middle Kingdom who were incredibly creative, ingenious, inspiring, and motivating. XW is none of those. Xiaodong typically speaks into the blackboard from only several inches away. His accented English is very good, except for his tendency to pronounce "delta" as "derta." This is clearly not a deal-breaker, but you may also find his tendency to say "this is just high school math," "nothing more than...", and "Okay? Alright" all extremely infuriating. He writes extremely fast, and only uses two board panels, so that he has to erase material unnecessarily quickly. I recommend that you make jokes (ideally non-disruptive) in class and ask questions to lighten the mood and encourage a modicum of Socratic dialogue. My last criticism of XW's teaching style is that he just shows up to class on time, stares at the class until people slowly hush up, teaches class for 75 minutes straight, and then leaves. He does not bother to show up to exams, and is sometimes condescending and dismissive at office hours. However, Wang Laoshi does wear stylish Puma indoor soccer shoes...check it out if you need to be entertained mid-semester. As for the positive parts of the course: this stuff is really useful. Fourier and LaPlace transforms at first seem highly technical and complex in the very dense Lathi textbook, but once you struggle through the problem sets, and read the text in the soberest of moments, you will come to see the light. In fact, the last class of the semester didn't cover any examples, but did a fly-through at Mach speed of all of the course material, from basic imaginary numbers all the way to LaPlace-generated block-diagrams of how to build highly-specified filters. The beauty in which this course merges with Circuit Analysis at the end is both incredible and rewarding. It makes all of the tedious, dense, analytic math seem worth the squeeze. In comparison to a substitute instructor (while XW supposedly spent three class-sessions in China at a conference), XW is way more organized (we did still manage to joke that we hoped XW would be barred from re-entry at JFK Immigration at least until the final). That said, the Greek substitute had far more of a sense of humor and presented the material in a far less rote, dry way. That said, when you're going through your notes before a MT or exam, XW's organization pays off. My final comment about this course is that many Asian foreigners (sadly not Asians born in this country, for the most part) seem to have been trained in some of this material as early as high school. Sadly we westerners are mostly left out. Thus, this course requires a bit of a jump in acuity from standard (advanced) Calc or Stats or Intro Circuits classes. But once you take this class, you will feel far more confident about rocking Stats or Intermediate-to-Advanced Econ. Just prepare to suffer unnecessarily thanks to WANG Xiaodong, since he seems to assume a lot of knowledge and acuity of his students (almost to the extent that he thinks they have seen all the material before). The exams are tough but not impossible. Prepare a good (permitted) cheat-sheet with as many formulae as you can cram in. It will pay off. Apparently this course was executed this year much better than last: more balanced and less challenging exams, and better feedback from TAs. That said, I found myself wishing desperately on multiple occasions that Vallancourt or Zukowski could have taught us the material. NighTalon, NYC, 12/2011

Aug 2011

I completely agree with the last review. The class was absolute hell, I hated every bit of it. He CANNOT TEACH, and grades badly - not the best mix of qualities. The class was terribly dry, so I just stopped going at one point. But make sure you get the notes, because they are important for the midterms and final (which was curved to a C+ or B-, wonderful). If you cannot get the notes from anyone then just go to class and suffer through it, but just write down whatever he puts on the board. When I finally studied the material, I found that it was actually kinda interesting, although challenging.

Dec 2010

Take a professor who lacks any teaching ability and would have difficulty explaining to a 4 year old how to add 2 +2, combine with a book that is completely useless, consider that this subject is fairly challenging to begin with, mix in a first midterm that is fair but a second midterm that is SO difficult that entire class bursts into laughter when time is called because they weren't able to even begin half of the problems, and as icing on the cake, put in a final grade distribution with the average grade awarded a C+, 17 % of students receiving a C- and 15 % of students receiving a D (yes, he gave 8 students a D), and you have Signals and Systems with Professor Wang. If you don't need to take this class, DON"T take it, you will not learn anything and it will only hurt your gpa. If this is a required class for you, my heart goes out. Pray every night that a different teacher might be leading it by the time you have to take it. And by the way, he distributed grades out on Christmas, just to make us all have a great day.

Dec 2009

Let's get the bad news out of the way: 1. He's HEARTLESS. Do not expect any concilation from him. If you made an honest mistake and wrote on both sides of your formula sheet instead of only one, he will give you a 0, and will announce it to the class. If you plead him to count the final more, he'll just say "No, I'm not changing the grade breakdown." If you ask him if it's at all possible to pass the class, he'll reply with "You know the grade breakdown, go calculate your grade." He will also tell you that he does not curve, that he will not answer any question regarding what grade you deserved. 2. He thinks we're RETARDED. If you ask him some question that he considers dumb in class, he'll stare at you for a while, then elaborate by showing 2 more steps, and then say "If you don't understand how I went from here to here, drop the course." 3. The TA is a Chinese PhD student. That means unless you can speak Chinese to the TA, he's not very nice, and he'll hardly understand what you mean. Now the good news: 1. Tough guy has a soft side. On the last day of class, he will tell you that your final will override the midterm grade ON THE LAST DAY OF CLASS! So if you bomb the midterm, don't be alarmed -- you will get a second chance. He DOES curve. His curve is weird though. He calculates the grade that everyone is supposed to get, and then he shifts everyone's grade up. My real grade was 2/3 of a letter grade higher than the straight out average according to the grade breakdown So don't believe the bullshit he tells you. He's trying to get you to study harder. Don't be discouraged. 2. Xiaodong Wang is a great instructor. His organization is crazy good. His notes are very good, and he doesn't ask anything on the final that he doesn't cover in class. If you understand everything he said in class (easier said than done) then you're fine. 3. The TA is a Chinese PhD Student. Also a good news, because if you speak Chinese to him, he'll love you. If you're among the lucky ones, then go suck up to him like crazy. He'll try to do everything he can to help you out.

Dec 2005

Wang is a terrible lecturer, doesnt know the material very well and simply reads off of pretty well made powerpoint slides. The amount you will learn from going to lecture is considerably disproportional to the 3 hours that you have to sit there listening to him. If you are planning to go, bring something else to do, unless you are planning on falling asleep. The schedule for the class was also very annoying, he ended up changing the dates of the first midterm, setting a new date for the second as a result and then changing that one as well. The material in this class is quite easy to grasp and can easily be done the night before a midterm or final. Homeworks are quite long and challenging at times. The course covers alot but there is only so much you can be tested on because alot of it requires a computer to do. It also becomes a lot easier if you come into it with a biology background. Most redeeming factor about Wang is that he loves to give out good grades and will try to maximize your final grade as best he can.

May 2005

Wang isn't a particularly good lecturer. He has a thick accent and just about all of his info comes straight out of the book--personally, I definitely didn't go to lecture very often. The material in the class, however, is kind of interesting, especially if you're good at math. It's pretty important stuff for a lot of the engineering applications. Wang uses a pretty complicated curve for his grading system, but the best part of the entire system is that if your grade on the final is higher than your average of the two midterms, your final counts as 80% of your grade and the midterms don't count. Wang's nothing special, but he's organized and you can always just follow from the book. Tests are challenging, but with pretty low means, you'll end up doing fine. I bombed both midterms, but nailed the final and ended up with an A-, despite going to lecture on only a semi-regular basis.

Sep 2004

Wang is a good professor. If you go to class, you will do well in his class. He provides all the information you need in class. He is also very reasonable with grading; and wants you to do well.

Dec 2003

Excellent professor. He has a good approach to deliver the materials. Lectures are very well organized. The exams are easy. most exam questions are taken from the homeworks, so make sure you go through all the homework before going to the tests.