David Jacobs

This professor has taught: CMSC422, CMSC426, CMSC733, CMSC798, CMSC828B, CMSC828J, CMSC828L, CMSC828P, ENEE899

Sort By:

Information	Review
David Jacobs CMSC422 Expecting an A Anonymous 12/28/2021	Generally a good professor for ML. Pretty responsive to student questions and very helpful/accessible especially during OHs. Overall, the class covered what I thought to be a pretty good range of different topics in ML appropriate for an introductory course, providing a solid foundation on the fundamentals including multivar calc, inner product, supervised learning, bayesian reasoning, and neural networks. Also, there were lectures on special topics(eg. GANs, transformers, CV, ect) that were really interesting. Although, the class/lectures did feel a bit unstructured/disconnected at times due to 2 professors coteaching the course. Also, especially for those with a background in ML, the class material did feel a bit shallow and didn't rigorously cover some of the algorithms eg. SVMs to the level I would have liked But overall, Prof Jacobs is a pretty solid choice and CMSC422 does offer what I think to be a good primer into the ML field that can be easily built upon through more advanced ML/DL courses.
David Jacobs CMSC422 Expecting an A Anonymous 12/25/2021	I don't usually do reviews, but I really wish someone would have warned me about this class before I took it. It is quite possibly one of the worst taught classes I have ever taken at UMD, mostly because of how its run. 1. Class Structure - This class is the most disorganized class in the CS department. Jacobs and Duraiswami co-teach, so even if you are in Jacobs's section you'll still have to deal with lectures from Duraiswami (and vice versa if you're in the other section). For some reason the professors thought it was okay to essentially add another "period" of mandatory content. We had in-person lecture on Tuesday, and Thursdays were reserved for "discussion" where we would do actual practice problems instead of theory. However, before the next week's Tuesday lecture, you were expected to watch the recorded lecture from the other section. So basically, even though you only registered for two 1 hr 15 min periods, you effectively had to complete 3 to make the most out of the class. Another issue is that none of the resources are centralized. They post the jamboard from lecture in the resources section of piazza, but discussion problems and jamboards are lost within the sea of piazza questions. Some of them were missing, and in general it was extremely difficult to access the content in a systematic way. 2. Content - I ended up self teaching a lot of the content in this class. During lectures, both professors tended to ramble and never really seemed to hit the point. Rather than using structured slides, they taught freely while writing things on a Google Jamboard (essentially a virtual whiteboard). This is fine if you know exactly what you want to talk about, but unfortunately Jacobs had a tendency to make a lot of mistakes and get confused while teaching. I never retained anything from lecture until I googled things on my own to learn them. I also felt like the general flow of topics could have been a little better. 3. Responsiveness & Grading- This class had one of the worst Piazza response rates. There were times where I would have questions that went over 4 days unanswered. This was especially problematic when a project was due the same week. A lot of times responses were also not that helpful. For example, I remember someone asked for clarification on one of the questions from the final review, and a professor responded with "I don't see what's unclear with this solution," and then did not respond to the student's followup. The grading in the class was also extremely slow. Problem sets and quizzes from the middle of october were only graded in the last couple weeks of the semester. Solutions were not released until the last week either, which made it very difficult to study or even know where you stand in the class before the withdraw deadline. I understand that grading of written problem sets takes time, but in general it felt like there was no communication between instructors and TAs. Overall I was extremely disappointed with the class, especially given that I love machine learning. Some pros are that the professors were nice and generally approachable. There was also a decent curve at the end which helped boost a lot of grades. I'm sure both Jacobs and Duraiswami are smart people, but I think they did a pretty bad job teaching. I would take 422 with different professors if you can.
David Jacobs CMSC422 Expecting a B- Anonymous 12/23/2021	I took this course in Fall 2021 Pros: + Generally nice guy + Pretty good curve + Pretty good lecturer Cons: - Some lectures, especially towards the end, seemed irrelevant to the course, only for them to appear on the final. This seemed unfair, especially when it was alluded that the topics were auxiliary to the course - Lecture notes are not easily available in one place, they were on Piazza but not readily organized, and half of them weren't even there. - Lecture barely helps with coding assignments - THE CLASS IS VERY DISORGANIZED! Along with what I've already said, lectures were delivered during the Tuesday morning section and the Thursday afternoon section. However, since you are only scheduled for either the morning or afternoon section, you would either have to find some way to get there in time, or (most likely), watch the lecture videos which are extremely long. They did not make it clear how exactly content was split up between both of those classes, or what quizzes were on what concepts, and it took me about a month and a half to figure out how the class was organized. - Workload is very high, even for a 3 credit CS class
David Jacobs CMSC422 Expecting a B Anonymous 12/21/2019	Friendly and knowledgeable. Good curve at end. Good intro to ML concepts, as it is not too thorough and covers a good amount of material.
David Jacobs CMSC426 Expecting an A SomebodiesGottaDoIt 12/12/2012	Good professor, good class. It's not so much about image processing as it is about computer vision, which turns out to be much more interesting. Dr. Jacobs is very familiar with the material and makes it easy to understand. The coursework is not very hard, but Matlab can be frustrating sometimes.
David Jacobs CMSC131 Expecting an A Anonymous 05/14/2011	Dr. Jacobs is a good professor, but he is new at teaching the 100 level classes, so he does not know how to really interact with students that have a very basic understanding of Java, or computing, in general. I came into the class with a firm foundation, and received an 95% in the class and got almost perfect scores on all of the projects, but I don't think I learned as much as if i were in another teacher's class. He works with other professors and so his exams and all of the material come from them, so the students are almost getting it second hand. He knows the material and is very friendly and a good person all around, but i am not sure that he is the best teacher to teach the lower-level undergraduates. If you can take another professor's class, like Nelson's, then it might be more worth-while and entertaining, but he is not a bad teacher, you just will have to be willing to jump-in in the middle of lecture and say "what are you talking about" at some point.
David Jacobs CMSC131 Expecting a B+ Anonymous 04/28/2011	Professor Jacobs is a nice guy and a good teacher. He covers all of the material that needs to be and try will answer any questions you may have; the emphasis of the previous statement is on "try" - sometimes he clarifies the thing you asked about, other times he leaves you more confused than before you asked the question. Overall, he's still a good pick for 131.
David Jacobs CMSC427 Expecting an A Andrei 12/16/2007	Good professor, definitely knows his material. Goes a bit fast, so if you can't keep up, definitely ask questions - he likes questions.