Chevron Left
Practical Reinforcement Learning(으)로 돌아가기

국립 연구 고등 경제 대학의 Practical Reinforcement Learning 학습자 리뷰 및 피드백

4.2
별점
365개의 평가
104개의 리뷰

강좌 소개

Welcome to the Reinforcement Learning course. Here you will find out about: - foundations of RL methods: value/policy iteration, q-learning, policy gradient, etc. --- with math & batteries included - using deep neural networks for RL tasks --- also known as "the hype train" - state of the art RL algorithms --- and how to apply duct tape to them for practical problems. - and, of course, teaching your neural network to play games --- because that's what everyone thinks RL is about. We'll also use it for seq2seq and contextual bandits. Jump in. It's gonna be fun! Do you have technical problems? Write to us: coursera@hse.ru...

최상위 리뷰

AK

May 28, 2019

This is one of the Best Course available on Reinforcement Learning. I have gone through various study material but the depth and practical knowledge given in the course is awesome.

FZ

Feb 14, 2019

A great course with very practical assignments to help you learn how to implement RL algorithms. But it also has some stupid quiz questions which makes you feel confusing.

필터링 기준:

Practical Reinforcement Learning의 104개 리뷰 중 1~25

교육 기관: Hamed N

Apr 23, 2019

I would give it -5 star if it was possible. The course material is so vague but still understandable if you sleep on them 10 times more than watching it. Maybe Andrew Ng courses or Python Course or Advanced ML course on google cloud (GCD ) spoiled me However statistically and self-judgement , this is not the case.

The instructor talking super fast and not understandable that could beat any translator machine I bet. What s more, the instructor talking about things which are not consistent with slides and also sometimes he does not explain some formulas or modelings.

The assignments are full of grammatical errors and they are super confusing. Very simple but super confusing leads you to have the grader failed you.

But , The worst part is if you take this course you will be all on your own and no body help you out as TA . If you check the forum discussion you see how many people complaining and how many questions left with no answer. I took this course as granted , but this is my responsibility to give back my feed back to potential learners.

Note that this is my feeling from the first week of class , I hope my idea change later.

교육 기관: Pedro L A V

Nov 27, 2018

Pros:

-It is a pioneer RL course in Coursera.

-Great exercise templates with interesting applications of RL algorithms.

-There are always references to good papers and new developments in RL.

-Good sense of humor in the lecture and templates.

-The discussion forum addresses the the bugs of the course.

-The course is challenging in the right level.

Cons:

-The lectures are not in that level yet ... they do not explain the important parts in detail.

-The lecturers should improve their public speaking and storytelling skills.

-The course subverts the sequence of the RL topics (cross-entropy is the first method and the multi-armed bandits setting is in the last week). This could be good, but ended up being confusing.

-The quizzes and exercises still contain many bugs.

Overall:

This is a good course, but it has the potential to be much better. If you want to challenge yourself and solve really interesting problems, take this course. You will probably have to watch David Silver's lectures on YouTube and read some parts of Sutton and Barto's book to understand the concepts. However, if you feel frustrated dealing with bugs in the exercises or answering quizzes that are confusing, do not take this course.

교육 기관: Xiao M

Aug 19, 2018

have to give a one star on this course, content hard to understand, speaker speaks too fast, programming assignment many mistakes, move on to david silver's youtube video for RL.

교육 기관: Jay G

Oct 30, 2018

Course 4 of Advanced Machine Learning, Practical Reinforcement Learning, is harder than Course 1, Introduction to Deep Learning. (I jumped to Course 4 after Course 1). That is saying quite a lot because I would describe Course 1 as "fiendishly difficult".

There's a few reasons for why 4 is harder than 1.

One big reason is, the course is still "in beta". Not everything, and maybe not anything, works as a straightforward Coursera Notebook. My workaround was to download the courses as IPYNB files, and then upload them to Google Colab. I'm glad for the experience as I'm now very familiar with Google Colab and how to navigate a Coursera notebook environment to get at the grader.py and submit.py files needed.

If you are not at least somewhat skilled as a programmer, you may want to avoid this course until it is out of beta.

Second reason is the Quizzes. These quizzes, most of them, are difficult. I myself never resorted to "try every possible permutation" to pass a quiz, but I did have to retake quizzes, re-watch videos, Ctrl-F find words in the video Interactive Transcript area, and read the Forums for help. Get ready to have some "fun" (and by "fun" I mean the opposite of "fun") taking these quizzes.

Third reason is, Alexander Panin can occasionally be difficult to understand in English (that's as gently as I can put it). But this, too, I'm glad for the experience. The neural networks in my brain for translating "thick Russian accent" to "colloquial English" have improved greatly. But everyone should take it easy on Alexander, because...

This course of his is awesome! I dreaded the Videos. I hated the Quizzes. And the assignments? Until I had finished an assignment 100%, it was the bane of my existence. But when you solve the assignment? Exhilerating. The assignments are a treasure trove of HOW-TOs on different RL techniques. Have you got an RL problem you want to solve? Chances are at least one of these notebooks will either flat out give you the solution, or else at least point the way forward.

교육 기관: MASSON

Apr 07, 2019

Interesting topic, however several things are not acceptable for a paid course:

+ Some assignments are a mess, it's crazy hard to get the environments working right, very little instructions and explanations

+ Assignment graders are broken and require you to fix them manually

+ No consistency between the notations of the different lecturers

+ Slides from videos are not provided (seriously ?!)

Overall, the course does not look serious, a kind of alpha version.

교육 기관: Tomas L

Dec 28, 2018

Still needs a lot of work

교육 기관: Kota M

Oct 04, 2018

The class is very immature as of September 2018. A good reason for taking this course is because it is one of few online courses where you can play with actual programming exercises of various reinforcement learning techniques, from dynamic programming to deep Q networks and actor critiques. Examples are mostly for environments of Open AI gym. You can also see examples where you use libraries such as tensorflow and pytorch used in the framework. However, the codes, including submission and grading system, have numerous bugs, which forces you to do extra debugging works unrelated to the course topics. Fortunately some early takers of the class left helpful comments on the forum, with which you can solve the most of issues if you read them carefully.

Quality of presentation is not as good as other courses I found in the Coursera. Most of the time, the lecturer seems to be just reading the scripts. To make it worse, the scripts are not written in spoken language.

교육 기관: Ajay K

May 28, 2019

This is one of the Best Course available on Reinforcement Learning. I have gone through various study material but the depth and practical knowledge given in the course is awesome.

교육 기관: Fan Z

Feb 14, 2019

A great course with very practical assignments to help you learn how to implement RL algorithms. But it also has some stupid quiz questions which makes you feel confusing.

교육 기관: Luke J

Oct 07, 2019

Challenging (unlike many other courses on Coursera, it does not baby you and does not seem to be targeting as high a pass rate as possible), but very very rewarding.

교육 기관: Roman P

Nov 05, 2018

The course is really in 'beta' state. Be prepared to struggle against not only the practical assignments themselves, but also against their bugs and assignment grading infrastructure problems.

But the course content itself is very useful and worth the trouble. Also, most of the bugs and problems are already solved by the community, you just need to visit the Discussion forums to find the solutions.

교육 기관: Simon V L

Oct 28, 2019

I've done about 14 courses on coursera and this was the worst. The teachers are so obsolete. They just rattle off a pre written text without any intonation. Instead of the videos it's easier to just read a book on reinforcement learning. I still gave it two stars because the programming exercises were interesting and usefull.

교육 기관: Zikai W

Jun 16, 2018

Indeed, this the 1st reinforcement learning course during May 2018. The topics and supporting materials are good for learning the course. Unfortunately, the course is not well-prepared in different aspects: 1) The assignments contained many bugs. One may spend half of the time to fix the bugs in the assignments. Sometimes, one may not be able to find tutor to ask for a help. The only thing one can do is helping herself or waiting for other classmates' feedbacks.2) Quiz is not designed for help one's learning. The questions in quiz are very confusing sometime. Also, one cannot get the correct answers after repeating the video several times. Sometime even one cannot find the topics in the lecture video. It takes you long time to try 'trail and error'.In all, it seem this course is not a well-prepared course in Coursera. I have paid and enrolled in many Coursera courses. Unfortunately, one might feel disappointed this time. A feedback from a PhD student (also a loyal customer of Coursera).

교육 기관: maciej.osinski

Nov 02, 2018

Brilliant content but quite some bugs in assignments

교육 기관: Mikhail V

May 23, 2019

The material covered in this course is very comprehensive, up-to-date, and broad. It goes far beyond typical RL courses/tutorials. BUT, at the moment the course is extremely raw:

1) For larger/longer assignment, it is impossible to work with coursera notebooks (keep disconnecting); It takes lost of efforts to set-up own environment (and you shouldn't really count on discussion forum for help).

2) The assignments have bugs / broken links and other issues.

3) Finally, I believe the main issue is that there is basically zero support from the course personnel/tutors. It looks like the course was just abandoned by their creators and they don't care about it anymore. Very sad, since the material is quite exciting and deep, and the course has lots of potential.

All in all: 5 stars for the content, 0 stars for the organization = rounding down to 2 overall.

교육 기관: Tingting X

Apr 22, 2019

I really like the lectures and homework, especially the coding assignments, which help me play games with RL and also improve understanding of the typical RL algorithms. Also, the discussion forum is very helpful and I can usually get out of stuck by following mentors' and other students' advice. Great thanks to Pavel Shvechikov and Alexander Panin for making such a useful course available!

교육 기관: Sahil J

Aug 03, 2018

Had a lot of fun doing this course. Although some of my fellow classmates are complaining that there are a few bugs in assignments, fixing those bugs itself can be a learning experience. The assignments,in general, are fun, particularly the honor's assignments.

교육 기관: Sergey F

Apr 09, 2020

At times it felt like a bit more video material would be helpful to better understand the subject/gain deeper understanding.

And fixing some of the notebooks would be helpful.

교육 기관: HSKim

Jan 29, 2020

Very practical lecture. I strongly recommend this lecture. Programming assignments are little difficult, but not impossible :) Just do it!

교육 기관: Vaibhav O

Mar 17, 2019

Well Prepared and taught course.. Will highly recommend as the primer for reinforcement learning

교육 기관: Thomas F

Aug 05, 2019

Course was very challenging what is good! Did several courses that were too easy. Quizzes were sometimes difficult to pass because of the way the answers are evaluated (all answers have to be correct) and even after watching the video several times the answers were not obvious.

Small things in the notebooks e.g. in mtc code was needed at a place but there was no comment saying that it is needed. In another notebook the wrong environment was loaded per default and had to be changed based on the notes given at the end of the page.

교육 기관: Chua R R

Dec 24, 2018

Great content! The python notebook submit problems leave a lot more to be desired.

교육 기관: Sergey

Oct 13, 2018

Доведите ноутбуки и grader до ума, не позорьтесь пожалуйста!

교육 기관: Keshav V J

Dec 27, 2018

This course was theoretically fulfilling, however i felt that the teachers failed to explain core principles with ease and felt a connection break in between their accent, their lectures and the slides in the background

교육 기관: Lars J I

May 14, 2020

The reading material was very weak and scattered. Some of the blog posts were nice but in general it's not very helpful to link a bunch of publications and a couple full-length books. Instead, It would be nice to have a small document that goes through the material in written form (if even just a summary).

The assignments were not very good and lacked depth. Often times I found myself implementing some formula/algorithm without knowing how it works or why. As such, it was easy to finish the assignments without learning anything. I would rather have two in-depth assignment (maybe an implementation from scratch with guidelines) than 6 or 7 shallow ones. The problem with the shallow ones is that there's no incentive to take the time and understand the functions that are already pre-coded. This makes it hard to follow whats going on "under the hood". You can, of course, still take the time to do this, but I feel like a true understanding of everything that goes on in all the algorithms in all the assignments would take far too long.

There was a general lack of theoretical material regarding the covered methods and algorithms. Why do they work?