Great!

great!

Great!

Good!

Good!

nice

good

good

Nice

I enjoyed this introduction course in Reinforcement Learning (RL). It explained in detail the fundamentals of RL such as k-armed bandits, Contextual Bandits and - of course - Markov Decision Processes (MDP). The lectures explained the conceps with nice examples and as well as the math behind (Bellman equations). The coursebook was the great "RL bible" ("Reinforcement Learning - An Introduction", 2nd edition by Sutton & Bartto); the lectures followed the first 4 chapters of the book quite closely.

I liked the programming assignments. It took some time to understand the structure of the tools used (e.g. the little known RLGlue) but after that it was quite straight forward, especially since the Notebook had great support for testing the solutions before submitting the assignment.

It was also interesting to see the guest lectures talk about the world outside the simple example MDPs used as examples, such as RL in the real world (using Contextual Bandits as a foundation), and about solving huge Fleet Management problems with RL.

One thing I missed in this course was more details about MDP and linear programming, which was mentioned in passim by the lecturers, and was an essential tool for solving the Fleet Management Problem (using approximate linear programming). Perhaps some of the next courses will discuss linear programming more...

I thought that the course content was extremely interesting, and the tests and programming were informative.

I did think though that the lectures were a little terse and could have given more information and worked through more examples. I think the presenters of this course and the people who constructed it could learn a lot from how, say, Andrew Ing's Coursera courses and Geoffrey Hinton's Coursera courses are put together and presented.

Specifically, the actual video time was very short and huge dependence was placed on the text book (which is very good textbook). I found Jupyter note book buggy and had to reset it a few times, but that might be me: I am not familar with it. I think as well, in a preliminary section, there could have been more on the Jupyter notebook and programming - even if this was just a document. As a user inexperienced with the Jupyter notebook, I found debugging and running test code in the lecturer's notebook in order to find my errors really hard. I often had to reset the notebook. Some assistance would have been appreciated here. In other courses that I have done, the prgramming environment has been more flexible which has made debugging easier, but I accept that my concerns here may be due to my inexperience.

Hi,

I don't know whether this feedback will reach the correct ears or not.

I have already completed the course before and now I am doing it again. One thing that I found is the coding assignments are using library and is not letting the student do the thing from scratch. Things will be very clear to the student if the build everything from scratch using the basic libraries. for eg. not using rl_glue, but coding up the environment, coding up the agent. Using abstraction is good, but for those who already know the things. Since this course is more about the fundamentals of RL, it should teach the basics of building environment, agent from scratch. Maybe we can use library once we have done it from scratch, like starting from week 3 or course 2. I persnally was not able to get the full understanding of the things untill I implemented the things from scratch.

Thanks:)

overall course very nice. A great effort !

The course is useful only if you read the book simultaneously. The book describes things in a much better way than the teachers of the course. If you want to get a good understanding of the concepts, I suggest you first read the topics from the book and then listen to the lectures.

Pros:

1. Good quizzes to understand the concepts.

2. Programming exercises are a must to understand the ideas correctly.

Cons:

1. One complaint I have for the course is, it lacks mathematical rigor in its explanations. For example, the course and the book just say that value updates eventually converge to the value corresponding to the policy, but they never reason about why that must be the case. Even putting a heuristic argument or atleast pointing to a "proof of convergence" would have satisfied the mathematician in me.

This course is not like the other courses where there are a lot of hand holding. This course is more of a ghetto mini course where they instruct you from where you should learn. The programming assignments and the graded tests then test your overall understanding. The four stars is because the topics could have been made a little more engaging. And I believe the programming assignments should have followed the overall philosophy of the course with minimal hand holding and requiring the students to build the classes from the scratch because the codes given in the assignments to assist the learners made the tasks a little less intuitive and less fruitful for people who are used to doing codes from the scratch.

This course is mostly about walking you through the first few chapters on the Sutton and Barto book, which is offered in a free pdf. You get some nice quizes and simple programming assignments and some nice animated graphics to go with the presentations. The exercises in the book are more challenging and open ended than in the course if you're interested in more work.

You don't need an extensive math background but must be comfortable parsing scary equations and be familiar with some probability/expected value ideas.

The videos are completely scripted which results in the lectures being a bit stiff and robotic. I prefer it when the professor just uses an outline and speaks from the heart.

Great course, I enjoyed it a lot. Though the lectures are kind of rigid in their style (reading from the teleprompter could be done by an avatar as well...), the literature is really good to learn from and there is some "bonus material" (talks of other experts in the field) that offers a focused glimpse into philosophy of posing problems, history of reinforcement learning and actual applications of dynamic programming. Beware of the short version of Warren Powell's talk in the 4th week: it was too compressed and gives you a bad impression of the lecturer, the long version is much better and easier to understand.

This course has a rich content to start reinforcement learning. The course book is very helpful. Assignments are very useful and effective to assess our knowledge. I have two relatively negative comments: 1) The verbal explanations during videos are very informative but not enough. Sometime a sentence that is said in 2 seconds require 10 minutes of work to understand. 2) The stuff team does not answer to the discussion questions, this might be improved. This course is my starting point to RL field. I find it very useful and thanks a lot to the team that creates this amazing content.

The course is overall very good, and it actually introduces you to Reinforcement Learning from scratch. Lectures are very clear, quizzes are challenging and the course relies on a text book, provided when you enroll. The only weak point, but not a serious issue, is that most of the lectures do not add content to what is in the book. Since studying the book is in fact mandatory, they could have used the lectures to better explain some concepts, assuming people read the book. Sometimes they do, but not so often.

I think the course's logic structure is so good. Even though I have read the textbook several times, I surprisingly found that many things I didn't understand or ignored when I was reading were clarified. The only thing I am not satisfied with is, the assignments (especially programming assignments) are too easy and the peer-graded assignment made me upset because I couldn't get my certificate immediately!

Overall, it is still an excellent course. Thanks for all your effort! ^_^

Overall fairly satisfied with this course.

Good coverage of the fundamentals through textbook backed up by videos and labs. Some of the quiz questions are a bit outside the box and include weird multi choice options that feel like they could be right depending on interpretation. I wasn't a fan of how the textbook handled Week 2 and 3, and spent a lot the time thinking "but why" - could be improved by explaining the policy and value dance from chapter 4 prior to commencing.

This course enabled me to be familiar with core concepts of Reinforecement learning. I was able to understand how Markov Decision Process and Dynamic Programming help to solve the problems. the lectures were clear and assignements were good and helpfull. I just expect to go more with how we can code agen-envirnoment interactions which are missed in this course. By the way, thanks for all the efforts done by the teachers.

Very good course which goes very well with reading the book alongside. I found very useful to read the chatper first and then brush and check my understanding by watching the videos. The explainations are clear and good and the videos length is just very good for me. Only thing I would improve is more coding assignment. With a more step by step series of exercises where one is learning to implement more things.

The course material (the textbook in particular) is great. I'm not sure how much value the videos add to the readings, but everyone has their preferred style of learning. My one dissatisfaction with this course is that I feel the material is not conducive to multiple choice quizzes. I wish there were fewer of those, and many more programming assignments. The coding parts were where I learned the most.

Great material! The course was very well taught and at an appropriate pace. I do think that the teaching style was a bit too formal, however. Also, the entire course, lectures, and order are centered around the book which is easy enough to understand on its own. It might be useful to discuss some practical tips and methods instead of only the book theory. Learned a lot anyway. Thank you!