Good complementary course to the Multivariate Calculus course by Imperial College London. I had already taken that course so I was able to skip almost all the lectures here, and jumped straight to the assignments.

Week 3 Newton's methods contained the use of Hessian instead of gradient descent, which was covered the the ICL course, but was not explained where it came from. They also used the hybrid of Hessian and Grad Descent and the use cases for both.

Overall, if you are stuck between this course or the ICL version, go with the ICL one

I also did the calculus course of imperial college london (didn't finish the last two weeks because the course gets worse after week 4) and I have to say that I enjoyed this course much more. Things are explained very clearly and I built a very good understanding and intuition of all the concepts. One could argue that the topics covered are a bit too basic, but in the end you get a good understanding of how optimization algorithms like gradient descent are working under the hood of ml-models.

Great Course! Here I find a lot of step-by-step explanations of how mathematical neural network works. A lot of examples and visualizations(a lot of animation, that can be used, as an experiment) provide a high understanding of what we using, why we using that in a certain situation, and how we can use it in practice programming neural networks. Highly Recommend!

thank you for recovering my knowledgement for math after so many years from pass my Uni

I find the teacher engaging and the idea that it is related to machine learning, but the labs are way to focused on machine learning, and i think they are poorly designed so that it is all busy work. Little things kept bothering me, such as how the quizzes pop up and block the slide that contains the information the quiz is asking about. You have to see both the quiz and the screen it is hiding at the same time. Then, I think the labs model poor programming practices such as packing data in and out of Python dictionaries all the time, and then my job is to go through and write code to pack and unpack the Python dictionaries. It is hard for me to write code I think is stupid, expecially when doing so teaches me nothing about either calculus or machine learning.

Since I got in touch with the field of ML/AI I was looking for a course to get a deep understanding of gradient descent. Finally this course made it happen and even more. My highest recommendations and thank you DeepLearning.AI for that great piece of work!

This course refresh my knowledge about calculus back in senior high school and even it makes me understand better about calculus and apply it in machine learning.

Calculus is a very difficult topic and yet the manner in which this course is delivered makes everything so very easy to understand. Incredible.

very very structured. Cant be more thankful to initiatives of Louis Serrano and Andrew NG, What a wonderful human service. Blessings from India

The content is fantastic, with a lot of illustrations and examples! Kudos to Luis Serrano and the DeepLearning AI team!

A good course for beginners. There are formulas that require a pen and paper to think and solve. Brilliant

Covers what is needed for ML. Provided examples, quizzes and labs are very helpful. Thank you

Those mathematical knowledge taught are very useful

There is a lot of useful knowledge

The best I have seen

So interesting!

Generally quite a good course, but occasionally the instructor glosses over important points or makes implicit assumptions that throw the learner off. One example: In the video “Hessians and concavity”, the instructor misleadingly states “Because all the eigenvalues of this matrix are positive numbers, then the function is concave up and the point (0,0) is a minimum.” But this isn’t strictly true. The function is a minimum if the Eigenvalues of the matrix are positive AND the point is a critical point i.e. has a zero first derivative. It's a pretty fundamental point and I wish the instructor had made that explicit. In addition to this kind of thing, there are quite a few verbal "typos" where the instructor says something that obviously contradicts what's on the slide. Would be nice if those were cleaned up.

Too simple, only the notebooks are somehow interesting

The course is meticulously organized into several modules that cover a range of essential calculus topics, including differentiation and multivariable calculus. Each module is designed to build on the previous one, ensuring a smooth learning curve for students of all levels. The course starts with a refresher on fundamental calculus concepts. This is particularly beneficial for those who may have a rusty background in mathematics. Detailed lessons on differentiation, including practical applications in machine learning, such as optimization and gradient descent. Advanced topics, including partial derivatives and multiple integrals, are crucial for understanding complex machine learning algorithms. One of the standout features of this course is its emphasis on practical applications. Each theoretical concept is tied to a specific use case in machine learning and data science. This approach not only reinforces learning but also demonstrates the real-world relevance of calculus in this rapidly growing field. Examples include: Optimization techniques in training machine learning models. Calculus-based approaches to understanding algorithms like backpropagation in neural networks. Applications in data analysis and probability theory. Interactive Learning and Assessments The course includes a variety of interactive elements to enhance learning: Whether you are a beginner looking to build a solid foundation or an experienced professional aiming to refresh and expand your knowledge, this course offers something for everyone. This review should provide a thorough overview of what prospective students can expect from the course, highlighting its strengths and the benefits it offers.

Very challenging for a beginner, but very well done! I had to research some of the concepts outside of this course, as I did not understand everything initially. I feel this touched on some very advanced calculus, and I personally needed more assistance (Khan academy, refresher of some algebra practices, etc). I also had to go through some of the videos a couple of times. However, I loved the Python labs! Very nice to apply the calculus we learned immediately in real programming examples. Definitely recommend for anyone interested in data science.

This is a great course to learn the application of calculus on ML. The instructor and his team have worked hard to explain important concepts such as the significance of derivative, partial derivative, how to find a minimum point of a function, Gradient Descent and Newton's method both analytically and graphically. The best part of the course is the Python programming labs. These labs are instrumental in developing a comprehensive idea about these concepts. I express my sincere gratitude to Coursera for offering such a beautiful course.

Excellent course to learn the basics of Calculus in a veeeery progressive way while acquiring essential fundamentals of machine learning itself like multivariate linear regression and classification applied to great examples of neuronal networks. Therefore, you don't only learn abstract mathematical concepts but also it immediate application to machine learning through a very well guided learning approach using well structured and pedagogic Labs and Programming Assignments

The course is quite essential for everyone who want to review math for machine learning. One outstanding point 's the course is intuitively all of theory. You can easily image derivative and the relationship between first derivative and second derivative. Moreover, you may learn the concept about machine learning and apply it to python code. Beside that, if you had experience about math, it may be too easier than you expect.

Clear and concise material, providing a nice introduction to understand the mechanics of the algorithms explained during the course. The labs are outstanding, providing enough context to complete them, without overwhelming the student, and at the same time showcasing really nice examples! Most importantly, it has motivated me to continue to study and expand my understanding of Neural Networks.

That was challenging, but doable! The course helps you understand the maths behind ML. Would totally recommend it to anyone who's preparing to enter the field of Data Science and Machine Learning. It took me longer than I expected though, at times needed to review some Khan Academy materials. The instructor is very good!