Bewertung und Feedback des Lernenden für Calculus and Optimization for Machine Learning von HSE University

I am sorry to point this out but this is course is my most frustrating Coursera expierience until now. The content is quite high level and requires a quite solid mathematical background. Unfortunately the lecturer's language and presentation capabilities are so poor that I found it really hard and sometimes impossible to follow the content.

This course is a great example, why people fail in math.

The teacher babbles stuff, which may sound simple for someone, who has studied the language of math extensively - yet he is talking to an audience, which didn't do.

The necessary vocabularies he introduces quickly in the "it's trivial" manner, which gave math professors already a bad reputation in university.

That combined with a strong accent and an uneasy way of talking, creates a very unpleasant experience.

I cannot even tell, if the material is difficult - I only can tell, that the teacher does a REALLY bad job teaching it.

very poor presentation

Sometimes it's hard to understand lecturer due to poor English (a lot of "Basically..." or "That's kind of nice.." is really annoying).

Atrocious experience with the answers parser in the mandatory quizzes.

its very hard to follow his lectures as well as his writeups and also its very difficult to understand his way of teaching and language.

This course has a lot of overhead. That is, it's too hard and time-consuming for what you learn.

For starters, it requires a refresher in algebra and trigonometry, e.g. quadratic equations, trigonometric identities, etc. since they are used extensively and the instructor just glosses over that.

The material is not always presented clearly. For example, after presenting a concept and the theory, instead of starting with a simple example and building from it, the instructor sometimes jumps directly into a rather complex one, and obviates some of the steps. The course has a lot of quizzes that require a lot of grunt work, and do not always help to solidify the understanding of the material. I found myself using other sources just to be able to really understand the material and pass the quizzes.

The final assignment on gradient descent is interesting, but again, there are a lot of details left out that require you to sift through the forum to understand.

Overall, if you are taking this class to get the certificate, it's doable and you will learn. But if you just need a multivariate calculus refresher for ML, there are more efficient ways to achieve that.

I found out the material is good but unfortunately the instructor often gives really bad explanation which bring me to somewhat lost during the course. But this course covers essential calculus that people need

First of all - I'm really sorry to tell you this, but Anton is an awful lecturer, and his English is really difficult to understand. He swallows words a lot, which is not great when you're watching math videos. I'm going through the whole specialization, and the first course was so much better, mostly because of the lecturers. In this one, I had to sometimes rewatch videos 3-4 times to understand anything, very happy to finish it and move forward to another course, where, I strongly hope, there will be no Anton. Again, I'm really sorry, but that is my impression.

The concepts are not explained clearly. The instructors English is too bad and its too hard to understand.

its very frustrating.

Explanations of the topics are not always clear, but sometimes they are. I think the teacher could do it better in respect to the explanations of the topics. This topic is important to understand optimization for data science and it would be nice to get more help in forums because most of the time questions take a lot to be answered. Last project was interesting because of the experiment.

A really disappointing experience. The content is weakly structured and the explanations are of low value. Everything feels rushed and the nothing is explained thoroughly. The whole feels like the team that made this course suffers from a case of the expert's blind spot. I left the specialization because of this course.

The professor mumbles and his english is lackluster. Very hard to follow

good course and lecturer.

it is nightmarish that there are no more his courses currently.I'll be waiting for his new courses.

good contents.

but somewhat challenging(especially for last programming assignment)

good peers(hints from forum is not a direct answer but very useful)

A really good course for refreshing basic math and understanding the reason why Gradient Descent is used

This course doesn't give you all the details, but makes you work hard enough to figure it all. 4.5/5

This was hands down the worst course I ever took via Coursera which go's to show that there is a great variability in Coursera courses, sometimes even within a specialization and one should check and choose each course carefully, before subscribing . Don't expect to actually learn something in this course. If you ALREADY know multivariable calculus and are proficient in basic Python (and preferably also Numpy , Mathplotlib and Pandas..) then you might gain some new perspectives and applications from this course. This will not be from listening to the lectures (which are concise, incoherent and incomprehensible..) but from the Quizzes and final project which will FORCE you to use the mathematics and the Python+libraries mentioned above. Unfortunately, quizzes and tasks suffer from language problems -poorly written English and badly formulated questions. Here's a small example from the final project which, is pretty representative of this course: In the final project one is asked to : "Write a function to compute the gradient of the Loss function in the given point". Seems quite reasonable at first glance until you look ,in vain, for the "given point".. Turns out there isn't one... After spending hours trying to figure out what I somehow, somewhere missed , I posted a message in the discussion forum and received the following response from a staff member (to the course's credit- there was one staff member who responded to all queries quite quickly and coherently.. a rarity in Coursera courses..)

"I think you are right, probably it is some kind of misprint, or maybe some math slang I am not familiar with"

So be prepared for similar "misprints" or "math slang" (Russian?) or just simple omissions of information which the authors of this course consider "obvious" and thus not worth the bother of writing down..

I found the course quite difficult (I have a physics background), but topics are well explained in the lectures and with help from the discussions I succeeded. I have really learned something :)

Very great content, once you get used to the way the teacher teaches, it goes really quickly

The system does not take correct answers due to not being written in a certain way, which confuses and makes that we spent a lot of time repeating the exams. The final project requires some knowledge of python and the guide is a bit confusing, so you need to support yourself from the forum.

Many of the subjects were not presented as clear as some of the subjects, in other courses for the same specialization

Informative but approach employed to teach material was somewhat abstract.

lectures were little off but concept wise it was good though it could be better

I purchased the course because of its great syllabus. I peeked in before purchasing, of course. My problem is that I came to this course to learn (not for the certificate), and I think this is more of a refresher of calculus rather than learning it from scratch. If you already know the topics and just got rusty after some time, then this is a great course. Otherwise, I recommend to think twice.

I would not blame the lecturer because the course may be rushed into production by the team. He is apparently great at math, but probably could do better at teaching it. Perhaps an update to the course videos would be great.

It was great to deal with this course as it helped me in gaining a much and important details and knowledge behind ML. Thanks a lot!