Overall, this course was very interesting and the organization made sense. However, if you are looking for a deep understanding of computational neuroscience as a computational novice, this course is much more advanced than they make it seem in the overview and it will not help. Conceptually, I feel I learned a lot, but practically, I still have no idea how to implement solutions. It is a must (not a bonus) to know MATLAB or Python coding. It is a must (not a bonus) to know calculus (definitely derivations, integrations, and limits in detail). The TA videos were one of the best parts of the course, but there is very little to no support if you are stuck with questions (the forums were not very helpful; get yourself a tutor in real-time). The biggest issue I have with this course is the lack of practice problems. I would suggest that at the end of each lesson, a few practice problems are given with full solution explanations. Then, a set of a couple of practice problems should be given to solve (with detailed solutions at the completion of the homework). Not until this practice and training are implemented should there be a quiz. The quizzes were the worst part of the course. They happen at the end of each week's lessons, and there are no real explanations for each solution. The forum isn't very helpful because it's a quiz, and they don't want to give out answers (understandably so). Therefore, a practice set should happen after each lesson but before taking the quiz.

It goes from too simple to very complex in few seconds.

Absolutely wonderful. It should be mentioned immediately for any readers who are interested in taking this course: While the course is itself "Introductory" and requires no background knowledge, a student with absolutely no exposure to introductory probability (Bayes theorem), programming concepts, problem-solving and differential equations may struggle with some of the more abstract concepts in the course. With that said, the rest of this review will describe my experience having taken this course with quite bit of prior work in Python, very little introductory knowledge of Calculus, and absolutely no exposure whatsoever to neuroscience or basic biology.

I was initially interested in taking this course to gain a deeper insight into some of the concepts I was introduced to on my regular visits with a neurostimulator patient at a local hospital. I am anything but disappointed by this course and feel that goal to be sufficiently achieved. Out of dozens of online courses I have taken on Coursera, this course felt like the most insightful and in-depth introduction to a topic I have ever taken.

The Computational Neuroscience course largely works to transition students from traditional thinking to thinking about extraordinarily abstract concepts. None of the problems or work in the course is truly 'difficult' beyond two or three challenging questions on the exams. Getting yourself to a point where you can think about these problems how you need to is the true 'difficulty' of this course. A student could, if they so choose, gloss over a large amount of the more difficult material to understand and still complete the course (i.e., one does not NEED to understand the derivations or proofs for, well, any of the equations given throughout the course) but would almost certainly struggle to complete a couple problems throughout the exams. With my insignificant prior exposure to some of these topics (and being absolutely incompetent with differential equations by my own admission) I would spend upwards of 8 to 12 hours per weekly quiz to truly understand the material, averaging about 15-18 hours a week to dive into the course material.

For students with little to no prior introduction to mathematics who are interested in taking this course: There are supplementary videos provided which serve as wonderful introductions to the concepts used for each week. While presented impressively well as far as a sudden introduction to upper-level concepts can be, they may fall short for individuals looking for a deep insight into some of the subjects throughout the course. As a bonus, I personally found the presenter for these supplemental videos to be absolutely hilarious.

Important to mention - Moreso than any other course I have ever seen on Coursera, the forums for Computational Neuroscience are active and filled with incredibly intelligent, kind, and helpful neuroscientists from around the world. The depth of discussion is absolutely incredible. The wonderful users on the forum served to provide nearly as much perspective and knowledge as the course itself did. I cannot emphasize in text how truly exciting it was to engage with such informative and interested peers and experts alike. Likewise, it's quite a pleasant and incredible experience to have the lecturers present an important concept in Neuroscience, with their own names on the citation/reference for that material.

For anyone looking to gain an insight into Computational Neuroscience and all the incredible technology that has arisen from the field: Would I recommend this course? If you're looking for something fairly rigorous and don't mind the potential for a bit of 'beating-your-head-against-a-wall' style thinking, absolutely! If you have absolutely no prior experience with programming or probability and would rather a less 'intense' insight into the field, this may not be for you.

This course felt really sketchy. Lectures were going way too fast and kept skipping concepts and derivations. Not bad if one only wants a superficial grasp of these concepts, but definitely not worth the price.

The first instructor is like Siri reading textbooks. Neither of the two instructors explained the concepts and calculations clearly. Just as another review says, they directly jump from too simple to too complicated. Overall this is one of the worst courses I have ever had.

This course is a good start for learning computational neuroscience. I have learned how a single neuron processing signal and how to modeling this interesting process through mathematical and computational methods. I also learned the idea and models of the adaption of synapsis and coding of a group of neurons, which form the basis of memory and learning. The course also gave a brief intro to reinforcement learning, unsupervised learning and supervised learning and I can't wait to explore the fascinating world of deep learning.

However, I think the course focusing a lot on the modeling of a single neuron rather than the modeling of a group of neurons. I think expanding the content and depth on neuron network would help the students have a better understanding on the memory or learning process. Also, the lectures in this course are very wide, whereas the quizes, although really help on understanding the lectures, are just related to some parts of the lectures. I think it would be better if the lectures closer relate to the quizes or give us a hint on where to look in the textbook in order to have a better understanding of the lectures.

Really appreciate all your efforts! learning how our brain works and the origin of our consciousness which I think is the ultimate topics of human being has always been my dream. I'm glad that this course enabled me to move closer to my dream

I'm a professor in psychiatry with a background in clinical psychology. I conduct clinical research to understand the neural mechanisms involved in psychiatric diseases. I found the course very informative and covers topics in computational neuroscience that are critical to further my research in the computational direction.

The course involves a moderate amount of math, which is absolutely necessary to understand the materials. For someone like me who did calculus more than 20 years ago (i.e., rusty), I often found the explanation of the math too fast. I had to pause the videos multiple times to digest the formulas and re-watch some videos to get a true understanding of the materials in order to complete the quizzes successfully (especially in later weeks as the concepts get more advanced). The supplementary tutorials by Rich Pang are extremely helpful. He talks at a slower pace, allowing time for you to think along the way. He is also very good at helping you to get an intuitive understanding of the complex concepts. I would recommend watching Rich's tutorials before watching the lecture videos. That way, you would understand the lectures more readily.

The quizzes are overall well designed and helpful in terms of facilitating the consolidation of your understanding of the concepts and methods covered in that particular week. I don't know if it's just me. I tended to spend a lot more time than the estimated time (e.g., 3 hours instead of 1 hour) to complete and pass a quiz (especially later ones that involve more Matlab programming).

Overall, I found this course very useful and overall well constructed.

Being a high school student with zero background in computer programming, i find this course a bit hard.

Nice content. Opens the gate of CNS for me.

But some explanations are just too virtual to be understood easily.

At last few weeks, the course talks about the relationship between NS and ML, which is astonishing.

I very much enjoyed the course overall. Lectures from week 2 to week 5 were a little bit tedious in my opinion, not because of the content, but due to the way the lecture was presented. I suggest that by the end of the course, one could see the correct answers (with explanation) of the quizzes, since that would help learning. I enjoyed the course and I learned a lot. I thank the coursera staff and the UW faculty who made this possible.

Fabulous course that helps students delve into the mechanics of the brain. Both instructors are amazing teachers. Towards the end ,the course also gives a flavour of DL & different paradigms of ML. However, i personally feel that although the course is very well structured, at times I felt that concepts were becoming too abstract and difficult to grasp.

A note of caution - This course requires good understanding of mathematics (Linear Algebra, Differential Eq, Probability and Statistics)

I think the course has many improvement opportunities:

-Some lectures just jump from too simple to too complicated skipping important concepts. They could take more time to explain topics some of us are not experts at (as circuits analysis).

-There are no mentors who answer questions in the forums, and there are some very common doubts. Apparently, there are even some mistakes in the tests, where non of the answers is correct.

-There are no explanations about the correct answers in the tests. There are many questions I still don’t understand why the correct answer is X or Y.

-Some questions in the tests have much higher level than the one of the lectures.

-For my taste, Prof. Fairhall goes too fast in lectures. She could separate them into shorter videos, where she widely explains more focused topics.

On the other hand, it is a very interesting course. I really enjoyed explanations by Rich Pang in the supplementary video tutorials, as well as the lectures of Prof. Rao.

Excellent course. Got to learn concepts across a wide range of subjects like information theory, statistics, biology, chemistry, etc. (I could go on).

Coming from a CS + ML background, this helped me appreciate the building blocks of abstractions that we can so easily take for granted in the age of Deep Learning. It really helps to learn and think about these things because it makes you realize how nothing is set in stone and the popularity of one model (MLPs) has a lot to do with history and not just mathematics.

The professors who teach the course are very engaging and are able to make a challenging topic into something interesting and entertaining. The course is very math and programming-heavy, so make sure to brush up on these concepts and be prepared to know how to conceptualize neuronal behavior with mathematical equations and programming functions and vice versa. If you have any questions to post on the Discussion Forum, be prepared to look for the answer outside of the course - the response times of the mentors or other students are horrendous and there are even some instances where questions are never answered. This, as a biology person who had little to no experience in programming or advanced calculus, was the most frustrating aspect of the course. In regards to computational neuroscience as a course, the material itself is beginner-level, but the math/programming is definitely not (more like intermediate/advanced).

The course started out quite well, but increasingly became very difficult to follow. Instruction during the second week through the fifth week was particularly difficult for me to understand. Only the additional lectures by Rich made sense, but didn't go deep enough to help me understand the course material.

Thanks a lot for a lovely course, I rushed through it in 2 weeks in excitement. This course is not only good in its content and rigor, I also found that I was able to absorb mathematical concepts much better than I would in a pure Math class. I typically took about an hour per 20min of lecture and paused in the middle to really take in the maths, and that helped a lot. The non-academic highlight of the course was definitely Prof. Rao sense of humor. Only thing that could be nicer is the Discussion Forum which was pretty empty :( All the more reason to join in and contribute!

This course is an excellent introduction to the field of computational neuroscience, with engaging lectures and interesting assignments that make learning the material easy.

Great course! Really enjoyed the variety of topics and the just enough computational work in the quiz's. And that Eigen hat had me smiling and laughing about it for a week.

This is a wonderful start for a biologist , to get idea of concepts of learning . An advanced course focused more on brain circuitry is suggested.

Thanks a lot

This course offers an introduction to computational neuroscience, which is rare in online education. I think highly of this. However, this course is not self-contained. To understand lectures thoroughly, you should search on the web and read textbooks and academic papers. I failed to finish this course at first when I had no background on computational neuroscience. But this course became manageable when I read a short textbook on computational neuroscience.