Professor Kao really, really cares about learning and is also a great lecturer, one of the best I've had at UCLA by far. I have a little bit of past experience with ML but Kao's slides and lectures made my understanding so much better, and the way the class is structured forces you to engage with the material. The final project is not bad at all and pretty easy to get 100% on if you can put in some time/thought -- even if you do it solo, IMO. The midterm was definitely very scary to me early on, but it's very similar to past midterms and the TAs do their best to help prepare you (I made some silly mistakes and still got an A on it). Finally, you do not need the prereqs to do well in this class as long as you're willing to put in some extra work early on.

Lectures are clear and slides are provided.

Deep learning results and certain concepts are interesting but theory is lacking.
Homework is pretty boring. It's all just working with matrix dimensions or finding gradients.
Get a good group for the project.
Midterm is difficult.

Professor Kao is the best professor that I have encountered. He has an unrivaled ability to interact with his students, fielding questions and reiterating content while making everybody feel included and listened to. Particularly, he is very good at noticing when he has lost the class and is willing to take his time to backtrack and ensure that the majority of students understand the material, and it is game changing. This class covers some of the most interesting and challenging material in the computer science world and professor Kao makes it all digestible while simultaneously moving at breakneck speed.

I cannot stress enough how uniquely good these lectures are. I have taken four machine learning classes and this one class taught me more than the other three classes combined.

The homework assignments are extremely challenging and very time consuming, especially if you are terrible at linear algebra like I am. However, they are very rewarding and forced me to actually learn the algorithms inside and out. The exam did a great job of testing your mastery of the material, and you are provided with a lot of practice exams and review material. I did not do well, and it was my own fault, and I still got an A in the class.

If you actually want to learn about machine learning, this is by far the best way to do it. I cannot recommend professor Kao or this class strongly enough.

This class can be renamed from Neural Networks to Neurological Damage due to the heavy workload. Homeworks take A LOT of time, and they don’t release solutions to the coding parts. Assignments build on each other so if you don’t finish one, you still have to work on it before starting the next one. Otherwise it’s time to watch your grade-ient descent down the drain! Kao is very smart and a great lecturer, and the TA's are awesome with their help during discussions/office hours and videos uploaded to Bruinlearn. The midterm was fair, but somewhat of a time crunch, and the project at the end is challenging, but they offer help and grade nicely.

Do not be fooled by the high A rate. This class is an unbelievable amount of work, definitely no walk in the park. Since I was taking it w/o pre-reqs, it was even harder for me. The one midterm exam was okay, I'd estimate about ~60% of the class our year got A's. 5 homeworks, with Homeworks 2, 4, and 5 being the hardest. 1 end of quarter project; I found it really annoying that the project was due at the start of finals week. This is manageable if you have a good group, but if not or if working alone, this will likely destroy any studying time you can get for your other classes. If you like ML or looking to fill your schedule, then you should take this class. Otherwise, would not recommend taking this class if you have a heavy courseload. The professor and TA team are pretty goated tho.

Kao a is an absolutely fantastic professor. His lectures are clear and engaging, and manage to break difficult concepts down into understandable chunks. He provides excellent slides, both annotated from class and unannotated originals, which are wonderful for studying. His slides often mention cutting-edge research in deep learning. Seriously, this is what a proper college class should feel like.

Although the class has listed prerequisites, they're not enforced. ECE 133A isn't really required (I didn't take it and did just fine). ECE/CS M146 isn't really necessary either, it's just background information that's mentioned in passing during lectures (I also hadn't taken it). You really do need to take a probability class though, even if it's not ECE 131A (STATS 100A or MATH 170E, etc. will do fine) or you'll be lost in the first half of the class.

The homeworks are quite time consuming, but there were only 5. They're a mixture of written math solutions and Python coding in Jupyter notebooks. It's helpful to have some exposure to Python before the class (even better if you already have familiarity with NumPy). The homeworks are pretty well spaced out, so there's plenty of time to complete them, and the TAs provide exceptional help during discussions (seriously, don't skip discussions. The TAs practically solve homework problems sometimes). Kao gives three "late days" across all the homework, so the deadlines are a little flexible.

Instead of a final, there is a final group project where you have to apply everything you learned in the quarter to a deep learning project. Kao provides a default project (in case you aren't creative, like me). It requires a fair amount of work, but it's due before finals week, so if you start early enough it doesn't interfere with studying for other classes. Getting a good group is essential.

Overall, this was one of the best courses I've taken at UCLA, and Kao is one of the best professors in the ECE department. If you're at all interested in machine learning, I highly recommend you take this class before you graduate. CS majors can probably petition it to count as an elective.

Professor Kao is super helpful and always willing to answer questions in class, during the break, after class, and in office hours. I personally did not take the pre-requisites for this class, so as long as you're willing to put in the time and effort, this class is great. Homeworks are quite challenging and vary in difficulty, but TAs were always willing to help, and a lot of people on Piazza had similar issues. The class is fairly math heavy, but basically everything is reviewed in the first 2 weeks, and you can spend extra time catching up on things you are a bit shakey on.
The midterm was very reasonable, and had an average of I believe 94%? Kao provided past year midterms, and I would say they were extremely representative of the actual midterm, so I was able to finish in the allotted time (with 30-45 minutes to double check my work).
The project was difficult in my opinion, since it's hard to know what architectures would work and perform well without actually implementing them and seeing how they perform, which is time consuming. Definitely try to get a good group so it's easier to distribute tasks and try various architectures.

the material taught in this class were definitely a lot more difficult than m146 or 145, he goes a lot more in depth into neural networks and cnn which is def helpful if you're looking to get an idea of what ML is rlly about. this class is very math heavy and i highly recommend taking m146 or 145 before this class.

jonathan was a great professor who rlly cared about his students and is great at explaining challenging concepts. the hws were challenging, but the TAs are rlly helpful and if ur rlly stuck, there is always github.

the midterm was challenging, but doable if you pay attention in class. the final project was graded relatively easily if the TAs can see that you put in an effort and achieve a better performance than the baseline.