Did not like how the concepts were introduced, it felt like learning theory for the sake of theory.

The course content is solid. The honours content is challenging and interesting. There's a couple of minor glitches that cause frustration in the PA's but nothing too earth-shattering. There's a lot of whining and whinging on the message boards, but take it with a grain of salt: the instructions to succeed in the programing assignments are complete and relatively simple, but you might have to dig around in lecture transcripts to put all the puzzle pieces together. The is GRADUATE LEVEL work, don't expect to be spoon-fed, and don't whine when you're not. I'd recommend the content to anyone. SO WHY ONLY 1 STAR? Because there is absolutely no support from TAs or Mentors anywhere. Nada. Zero. Zilch. They are asleep at the switch. If you expect any kind of interaction to expand your learning horizon then you will be sorely disappointed. I sure was. The lack of engagement from the TA/Mentor community takes what could have been a 5 star experience and drops it to zero. But I can't go that low, so 1 star it is.

This course is really amazing. The lecture is well-organised and lecture material is good. This course covers basic knowledge about representation in Probabilistic Graphical Model. It includes Markov Network, Bayesian Network, Template Model and some other knowledge. The assignments, oh, I have to say, although some quiz in it seems like having bug, are still impressive. I strongly recommend finishing all the programming assignments of this course. Some trick parts of the knowledge taught in the course are covered by the assignments (like template model part, trust me you have to think about the template model part really, really carefully to figure out what it exactly means). Anyway, it worth my payment :-).

If you wanna take this course, buying a textbook is a good choice because there are some extra knowledge which is not covered by this course in the textbook. However, without a textbook you can still continue. I really appreciate Professor Koller for offering such a great, amazing course!

In the video, a lot of knowledge point do not explain very clearly, we do not konw how to resolve the quizzes. Moreover, if buy the textbook, may acquire more detail about PGM, but the textbook do not explain very clear neither. Textbook is hard to read. Even so, this course is worthwile to learn. Because PGM is one of the basic theory of machine learning and widespread use. In the end, thank Koller and coursera! Thank you very much!

Dear Madam thanks a lot for the course.

This course - in addition to Machine Learning, by Andrew Ng Sir, are perhaps most comprehensive courses.

This course covers a lot over a period of 5 weeks. It demands higher level of focus. So, the learning still continues..

Regards,

Santosh Kumar Singh

Bangalore, India

The instructor provide clear explanations and useful examples.

Reading the recomended resourses, including the books, are also help me to comprehend the course contents.

Great course overall, thanks

Superb exposition. Makes me want to continue learning till the very end of this course. Very intuitive explanations. Plan to complete all courses offered in this specialization.

A comprehensive introduction and review of how to represent joint probability distributions as graphs and basic causal reasoning and decision making.

Very interesting and exciting course.

Thoroughly enjoyed the course, although some minor issues;

- No interaction on the discussion board, despite it being advertised as such

- No practise questions provided while learning. Pre and post questions available.

- Exam questions often feel out of order i.e. having a question about a topic for it to be discussed in the next section

- Final exam contained a question that was explored in the Honours section, meaning I had no clue how to answer it and couldn't get 100% on the exam

Great course! Lectures are clear and comprehensive. Quizzes really check knowledge and are challenging. In the programming assignments the main focus is put on implementation of PGM algorithms and not on technical aspects of Octave/Matlab. Some changes could be made in Programing Assignment 4 to make description and provided code easier to understand.

Lectures very good, but the code in the programming assignments is awful.

Having done the first few programming assignments, I decided to switch to recode and do the programming excercises in python/numpy/scipy etc.

The code definitely should get an update, especially because for instance tensorflow starts to integrate tensorflow probability.

Video lectures were clear and the course content was detailed and explained clearly. I take one star off because some of the material needed for the quizzes wasn't in the main course material; for example, the sum-product algorithm was mentoned in one of the quiz questions, but wasn't mentioned in the main material.

Everything is fine except the bugs in programming assignments. Although it says advance course, the programming assignments aren't that hard. The problems is difficult to submit it to Coursera.

Great course. Explained in a straightforward manner.

The course contents are presented very clearly. Difficult ideas are conveyed in a precise and convincing way. Despite this, the global structure is not presented very clearly, and the quality of some course material is not excellent. In particular, I didn't find the optional programming assignments particularly interesting, and the code/questions contained more than one bug. Also, the quality of video/sound is quite poor, and varies a lot from course to course.

The course content is really interesting and Daphne Koller is a fabulous presenter. Unfortunately, though, you are doing this course on your own - looks like there have been no TAs online for over 3 years, and if you're looking for support or assistance understanding any of the work you may find confusing or difficult then don't expect to get it here. Very disappointed that a paid course has virtually no support in it whatsoever.

Information is well presented. Tests are 4 questions. Any mistake in the answer counts as wrong, and all questions must be correct to receive the passing 80%. The course makes you wait an hour to retake the exam, so it is NOT friendly for folks on a time schedule.

Great topic, the professor is a top expert in the field, but the grading interface badly needs an upgrade. It is not acceptable for students to have to manually hack JSON submissions just to get around grader errors.

Quite difficult, not much help in discussion forums, some assignmnents had insufficient supporting material and explanations, challenging overall, I thought at least 3-4 times to abandon it.

The audio is VERY VERY poor.

That makes it very hard to understand what Prof Kohler is trying to impart on us..

I often lost track

This course should be not easy by itself, but the lecture is not organised, which makes the course very difficult.

Very informative and exciting course. The lectures could be better organized and quizzes made simpler.

Quite abstract. A solid mathematical grounding, but largely devoid of practicalities. Optional exercises are quite basic, and don't get to the heart of the matter. Lectures are confusing, as undefined terminology come up out of the blue, and key concepts aren't clearly explained.

This course is poorly structured, the material is poorly explained, the lecturer is going too fast and does not stress important concepts, video, and sound quality are below average. Do not recommend.

The structure of this course is an example of how not to teach mathematics. Examples before definitions and introduction of general concepts, lack of direction and "big picture" context, unexcusable things like "let's prove it by example"... It is very frustrating and almost impossible to follow.