| Information | Review |
|---|---|
|
Uzi Vishkin
ENEE459P Expecting an A Anonymous 12/20/2024 |
This course on parallel algorithms was very interesting, I would recommend it to others who are interested in the content and have sufficient background. The class provides an overview of parallel algorithmic theory with several useful algorithms presented. Note that this class is the undergrad section of ENEE651 and CMSC751, but is the exact same class with the same classwork and exams. Classwork consists of several theory homeworks (worth 10%), five programming assignments (worth 25%), first exam (25%), and a cumulative second exam (40%). Pros: Programming assignments are interesting, they provide a unique opportunity to write parallel code. Theory homeworks are fairly low-stakes being worth so little. Exams are generally fair, although there can be a bit of a time crunch. Algorithms are fairly well-explained and the professor is always happy to answer questions. Cons: Theory homeworks can be difficult and time-consuming despite being worth so little. Also, solutions are never released. You will need to know the solutions for the exams, so make sure to go ask if you don't know what you got wrong. Programming assignment difficulty varies wildly: the first assignment could be implemented in ten lines of code or something, the last was quite difficult. The course prerequisite should be updated in my opinion. It currently requires CMSC351, but the course occasionally assumes knowledge that not all 351 professors teach. If you intend to take this course, I would recommend either taking CMSC451 first or self-studying. The class is not offered every semester. Logistically this makes it somewhat difficult to take as an undergrad. On the whole, if this is a field that interests you then you should definitely try to take this class. |
|
Uzi Vishkin
ENEE651 Expecting a B+ Anonymous 12/19/2024 |
he is very intelligent but his presentation of parallel algorithms has not changed since the 80s. this is generally fine, since professor vishkin has a great deal of original research in parallel algorithms to his name, but the limits of the course show up in the details. everything is taught out of his personal notes, which are essentially just a review of his research career presented in an order he finds suitable for education. the notes are generally detailed with references, but they have not been updated in over a decade and are frequently vague and typeset poorly. whenever possible, i found it more useful to study from professor joseph jaja's textbook (also from UMD!) as the explanations, order, and presentation of algorithms was better dry homeworks, presented weekly, were 10% of the grade in the course. these are exercises from his notes, which are often meant to "fill in blanks" that aren't explicitly spelled out. programming assignments were 25% of the grade. i liked these; you design and run programs for a parallel computer that professor vishkin designed. the programming process is very much like overlaying the work-depth model onto C. they were not particularly hard except for the last one the exams in this course were awful. he expects you to print out and bring his 100-page notes to the exam as an exam aid. you are not allowed anything except an unmarked copy of his notes. if you do not print out the 100 pages of notes and bring them to the exam, you can expect to be tripped up by some tiny detail you never expected to be important and punished harshly for it. meanwhile, the mouth-breather next to you who printed and BOUND the notes will have a great time. ask me how i know! also, the second midterm is weighted much more heavily than the first while also covering more (and more difficult) content overall, i do feel like a learned a lot about parallel algorithms in this class. but the process was pretty excruciating. it's hard to stay engaged in lecture when lecture is just professor vishkin putting his notes up on the projector and reading through them to you. it's even harder to go back home and read through those same notes to study for the exam that is built entirely around those notes. i appreciated the practical programming components of the course, but i quickly grew tired of the professor constantly singing his own praises and reminding us how important he was to the development of parallel algorithmic theory. i can think of at least two faculty from the CS department who are equally qualified to teach this course and certainly would have done a much better job |
|
Uzi Vishkin
Anonymous 12/16/2018 |
If you are forced on gunpoint to take his course then I wish you gl. He's so sloppy when explaining |
|
Uzi Vishkin
ENEE446 Anonymous 06/14/2013 |
Lectures were dry and very text dense - think slides upon slides filled with 9 point font. There were about six chapters in the book, and each had its own 100-150 slide set provided by the publishers, which is exactly what was went over in class. Vishkin adds some notes here and there but it's more or less words words words. He skips very quickly over the book stuff in an attempt to get to his own research, which is fairly interesting, I suppose. He mentioned like 10-20 times over the semester he was researching so and so and we were encouraged to research under him. The book slides aren't fantastic for both this reason and the fact that they take, say, 10 pages to explain something that is in reality quite simple. The exercise are quite long, due about every three weeks, and they are annoying only because they range between very vague, open ended design questions to stupid common-sense questions. They were also not graded especially well by the TA. I had some of my solutions which were entirely incorrect get full credit and some solutions with a small typo get almost none. At a certain point in the semester I just started bullshitting the whole thing in an hour or two and getting the same scores as when I was spending 10 hours on each assignment. One test, one final, both straightforward. Some definitions, some simple calculations, and some open-ended thought questions. 75% of the class is more or less rehashed from the honors version of 350, so I pretty much got an A not studying at all and just remembering what I had already learned. |
|
Uzi Vishkin
ENEE446 Expecting a B jmistry3 05/23/2013 |
ENEE446 is a very insightful course, which in combination with Prof. Vishkin gets even more powerful. This course is not for the one's who just want to do exercise at the end of each chapter, but with this course students are expected to make equal effort as much as prof does on you. Prof. Vishkin has deep understanding and passion for the material required for this course and he delivers it at its best. The fact that he provides real life examples and current trends in computer industry makes his lecture very interesting. There's allot to absorb from the course and prof. Vishkin's lectures, so a constant effort interms of reading the material of textbook and attending the class lectures attentively is required. Final comment on this note is that my perspective towards pc have changed from what I had before taking this course and what it is after completion of this course, and am surely considering to make my career along this path. |
|
Uzi Vishkin
Expecting an A ArunVenkatesan 05/16/2012 |
I took the ENEE 759K[Parallel Algorithmics] course with Dr.Vishkin in Spring 2012 and I would definitely rate it as one of the best I have taken thus far at UMD, in terms of the depth in the concepts, the cutting edge nature of the field and its future scope. LECTURES: According to me, the most valuable part of the course were the lectures. The Professor clearly had a thorough understanding and passion for the field and he took care to ensure that the students understood the material taught in the lectures. COURSE MATERIAL: A key factor of the course is its breadth - the material has been chosen to provide the students with an exposure to a range of Elementary routines, Techniques and Paradigms in the field of Parallel Algorithms. The students are also taught how to apply these "tools" as building blocks to create more involved Parallel Algorithms. At the end of the course, with a good grasp of the various tools and their application, the student will be able to get a clear idea of how to approach any new problem in the field. ASSIGNMENTS: The programming assignments gave an opportunity for the students to actually implement the techniques learnt and produce very significant speed-ups. It also enabled the students to see for themselves that Parallel Programming on XMTC[PRAM-model] could be much simpler when compared to other Parallel Programming techniques. The Dry homework tested the ability of the students to apply the concepts learnt in class. EXAMS & GRADING: The two examinations were open-book and tested the student's understanding of the techniques. The Professor also gives due credit to students who demonstrate clear improvement as the class progresses. Both the Professor and the TA were very helpful and were quick to clear every doubt the students had. The students were also given a clear break-up of the points for the Assignments and the Exams. I would strongly recommend the Parallel Algorithmics course by Dr.Vishkin as it is very relevant and the concepts taught have significant scope for application in the future. |
|
Uzi Vishkin
ENEE641 Anonymous 12/22/2010 |
His homework is straight out of the posted solutions on the book website, and won't care if you copy the answers. Teaches well, but his exams are really tough, and his grading scale is very subjective - the final can be up to 85% of your grade, so you won't know your grade until after it's posted. DON'T take 641 unless you were a CE major - or you might have to drop it after getting back the midterm grade. His exam reviews are very short - he will not give you any hints or general areas to study, nor will he review material unless someone asks a specific question about it in advance. He also advertises his parallel algorithms class, which only like 4 people from MD registered for this past semester |
|
Uzi Vishkin
Expecting an A jamedw81 04/27/2010 |
I am surprised that so few have registered for the Fall 2010 ENEE459P course on parallel algorithms. I took ENEE459P with Dr. Vishkin in Spring 2008 and found it to be a relevant and informative course. The lectures (two 75-minute lectures per week) were very helpful in understanding the course material. The professor often asked students questions during the lecture to make sure everyone understood the material, and he answered any questions I had as well. I could tell that he cared about the material he taught and understood it well. There is a book-like set of class notes posted on the course website, but I found that the notes were easier to understand after attending the lectures. As a computer engineering major, I found the programming assignments to be a fun and valuable part of the course. The assignments are loosely based on algorithms given in the class notes, and the assignment sheets clearly explain what needs to be done. The assignments require only a little knowledge of C programming and relatively few lines of code. Most of the work is in understanding how to translate the high-level descriptions given in class into a working program. I found this to be the most educational part of the assignments, and it helped me to appreciate how simple parallel programming can be given the right platform. The written homework was more difficult but worth the trouble. The questions required some thinking beyond merely restating things from the text. I personally appreciated the challenge, but even if you have a hard time with the questions, it does not hurt your grade very much. Figuring out the answers to the questions definitely helped me to understand the subject. Grading was very fair, and the grader/TA explained why points were taken off for certain things. The professor and TA were available to answer questions outside of class. The professor was especially willing to spend some time with students to explain anything that they found confusing. Overall, I would definitely recommend this course to computer science and computer engineering majors, and would encourage anyone interested in the future of computing to consider the course as well. |
|
Uzi Vishkin
Expecting an A- Anonymous 03/28/2009 |
I was so excited when I registered for this course (ENEE159V). I am sophmore in Computer Science. The grade I expect is at least A- and if i get lower than this grade I would be mad. However, The lecture was so boring and I think the professor should divide it into two lectures per week. The notes were unclear and they were not for freshmen. they should be taught for seniors. This course is very difficult his projects are waste of time. The projects takes at least 20 hours and maybo more to be done because they were unclear and not related to the notes. If you want to understand and finish the project you should do research and read other books and article on the internet. The TA was unhelpfull. if you send him an email he doesn't reply. I think he was easy on grading. and the good thing about the class that he didn't have tests. if he would have tests i would fail the class because i didn't understand anything. The professor was good and nice with students. But he needs make it easier for students and to have more clear HomeWork and notes. |