I think there is actually very little you need to know on whether you should take this class. Know Python (if you do not, homework will take a lot more time than it should). Be familiar with manipulating arrays and lists and especially with numpy functions. You can definitely pick it up as you go, but it will cost some extra time. The homework are difficult sometimes, but Kao will give you everything you need to know to answer questions; if not him, the TA's. Kao is probably the best professor at UCLA and his lectures are actually the most engaging and inspiring things to listen to. He keeps the students engaged, answers any questions, but most importantly, he shows that he cares. He is not some professor that is pompously concerned about their research that they view teaching as a second priority. Kao shows that he cares about teaching and I think that is all the reason you need to take this class. It is a lot of work do not get me wrong (in fact he will tell you this before hand); be familiar with Linear Algebra and Probability and you will end up with an A if you do the work and understand the concepts.

Kao is one of the best professors at this school. He is clear, engaging, informative, and an incredibly supportive teacher. Would highly recommend both Kao and this class, and would easily choose to take this class again.
If you've had 102 with Kao, his style of teaching for this class is very similar. He posts unannotated slides before lecture, annotates them during lecture, and reposts them to CCLE afterwards. His unannotated slides contain information and videos, but most of the derivations and math he does by hand on blank slides. He uses polls to monitor class comprehension, and frequently stops to ask for and take questions. He uses piazza to allow students to answer each others questions, but both Kao and the TA's are present to resolve ongoing confusion.
The neuroscience and probability homeworks were written, while the decoding and classification were jupyter notebooks. They walk you through complex concepts in small increments, and are interesting and fun to work through (and great for learning python!).
Brain machine interfaces were one of the first things that drew my attention to electrical engineering, and I found it incredibly interesting to take a class in exactly that. If you have the opportunity, I would highly recommend taking this class.

This is the third class I've taken with Professor Kao, having taken his Systems and Signals, and Neural Networks and Deep Learning classes. As always, Kao's lectures are very clear, informative and interesting.

The HW can be a bit tricky- start ahead of time and go to office hours. Throughout the quarter, Kao was very accommodating and even relaxed the grading scales at the end of quarter.

Due to COVID-19 and remote learning, we weren't able to cover the amount of material the class usually covers. However, for context: the first third of the class covers basic neuroscience, including action potentials and how the brain works. The second third covers Poisson processes and discrete classification. The third half covers decoding including Wiener and Kalman filters, which I think is the most interesting part of the course.

If you are looking for an interesting and useful elective, this is the class for you. Highly recommended- 10/10.