Chevron Left
Back to A Complete Reinforcement Learning System (Capstone)

Learner Reviews & Feedback for A Complete Reinforcement Learning System (Capstone) by University of Alberta

4.7
stars
630 ratings

About the Course

In this final course, you will put together your knowledge from Courses 1, 2 and 3 to implement a complete RL solution to a problem. This capstone will let you see how each component---problem formulation, algorithm selection, parameter selection and representation design---fits together into a complete solution, and how to make appropriate choices when deploying RL in the real world. This project will require you to implement both the environment to stimulate your problem, and a control agent with Neural Network function approximation. In addition, you will conduct a scientific study of your learning system to develop your ability to assess the robustness of RL agents. To use RL in the real world, it is critical to (a) appropriately formalize the problem as an MDP, (b) select appropriate algorithms, (c ) identify what choices in your implementation will have large impacts on performance and (d) validate the expected behaviour of your algorithms. This capstone is valuable for anyone who is planning on using RL to solve real problems. To be successful in this course, you will need to have completed Courses 1, 2, and 3 of this Specialization or the equivalent. By the end of this course, you will be able to: Complete an RL solution to a problem, starting from problem formulation, appropriate algorithm selection and implementation and empirical study into the effectiveness of the solution....

Top reviews

JJ

Apr 27, 2020

This is the final chapter. It is one of the easiest and it was fun doing that lunar landing project. This specialisation is the best for a person taking baby steps in the reinforcement learning.

CR

Feb 26, 2020

Great course for learning the fundamentals. I liked that it tied into function approximation for deep reinforcement learning. The text book made the fundamental concepts more clear.

Filter by:

26 - 50 of 128 Reviews for A Complete Reinforcement Learning System (Capstone)

By Jesse W

Jul 29, 2020

This course ties everything in the previous three courses together to simulate a reinforcement learning system for landing a lunar module on the Moon. The programming assignments are more or less guided, as in the previous courses, and the capstone project doesn't take very long to finish, despite the nominal 6-week length. The end result is fun and satisfying. Overall, I would highly recommend this specialization to anyone who is curious about reinforcement learning methods in machine learning.

By Mukund C

Apr 2, 2020

Absolute fantastic!! Thank you to everyone that put this course together. I loved the "behind the scenes" decision making on choices of approaches - I just wish there were some more them to "distinguish" the process for making different choices. The instructors are excellent teachers. Would love the opportunity to sit in a class live and interact and ask questions!! Was great fun digging into the code and understanding the data structures.

By Cesar S

Sep 28, 2021

This course, and the whole specialization, have been transforming for my career, as I am willing to pursue a PhD on reinforcement learning as soon as I graduate. I am very thankful that this course exists, because it got me to understand from the fundamentals of the area to advanced topics in a digestible, well-structured, with lots of material to cover and challenges to overcome. Thank you so much!

By Akash B

Dec 8, 2019

This capstone project is really amazing as how it gives the overall expertise understanding for experimentation and how to implement the algorithm. From MDP to scientific selection of meta-parameters are really important to decide about how should be make an agent, but there are lots of considerations.Overall, this was a great experience and would remember the instructors for all my life. Thanks.

By Walter O A

Jan 18, 2020

A solid introduction to the subject matter of Reinforcement Learning. Especially helpful navigating through the Sutton & Barto book. The programming labs all worked and included robust tests for correctness. I especially appreciated this as I have spent significant time in other courses banging my head on the wall because of an incorrect or vague lab assignment. Well worth the time invested.

By Varun B

Sep 20, 2020

The instructors are simply amazing! They cover a lot of algorithms, highlighting their motivations, use cases, and limitations. What fascinated me is that all the different topics covered are based on very simple, yet powerful ideas. Because of this course, I feel equipped with knowledge and experience to start working on my own projects and consider pursuing RL and AI for my graduate studies.

By Pavel I

Jul 27, 2021

Specialisation is really worth it. Especially it is really great introduction for beginners in RL. Martha is great instructor, all her lessons were really clear for me and easy to understand. Adam's lectures as for me require more time and effort to catch sense, but still really good. Quality of every course in that specialisation is higher than average on coursera I definitely recommend it.

By Dale G

Aug 2, 2021

Excellent course and specialisation for those who wish to understand the basics of reinforcement learning. A must complete specialization to those who are interested in the Reinforcement Learning: An Introduction by Sutton and Barto as it greatly enhanced my understanding of the content with videos, quizzes, and programming in addition to simplified examples of the core concepts.

By Francois L

Dec 21, 2020

It goes without saying, this was a fantastic introduction to reinforcement learning. There is so much to learn though, there is room for the authors (or any other author) for the creation of more specialized classes, or conceptually centric ones (why does RL work at all). In short, hats off!

By Niju M N

Oct 26, 2020

The Final project is Lunar Lander , applying what we learned in the previous courses in the specialisation.The assignment is friendly with more comments and easy to understand. It covers the implementation of Adam optimiser too .Over all a good course and specialisation

By Lim G

May 14, 2020

This course sums up all the fundamental knowledge acquired from the previous 3 courses. It covers comprehensively on a sample end-to-end RL problem think process, application and assessment on the performance of model. It helped me better visualize how RL can be applied.

By Jau-Jie Y

Jul 7, 2021

Thanks to Prof Martha White and Prof Adam White, for their hard working on teaching.

The Prof Drew Bagnell lecture on "system ID and optimal control" is interesting. The Prof Sucan Murphy lecture on "RL in mobile health" is also interesting.

By Arnold S

Jul 27, 2022

The course material is clearly explained in logical steps to build intuition. After learning multiple complex but simple to understand aspects in the first courses, it all comes together in the later courses of the specialization.

By Rafael B M

Sep 1, 2020

This capstone works as a case study of the application of the full RL framework into solving a complex control problem, in this course, it's possible to review and reinforce all the knowledge acquired from the previous courses.

By Steven H

Jul 12, 2020

Would be great if we can see the video of our agent instead of a pre-recorded video. Also, the checking is not perfect and inconsistent. A great course and I learned a lot in RL. Thanks Martha and Adam! Wish you all the best!

By Lars d R

Apr 17, 2020

Best course I have taken on Coursera so far. A great mathematically oriented introduction to RL techniques, ending with the modern state of the field. I would reccomend this course to anyone interested in RL!

By Bei Z

Feb 28, 2021

Step by step to build a RL system and instructions on what needs to be considered in each step. Those sessions by experts are really good. all of them are instructive and give view of RL usage in real world.

By Anas F

May 30, 2022

Excellent course and excellent labs. The course is so well organized and well explained that it feels like a story with chapters. I wish there was a pdf with all the slides for future references. Thanks!

By George M

Apr 27, 2021

Very good project!

It was interesting to put it all together.

The videos guide you through the thinking process to understand the limitations and features of the project before concluding with the design.

By John J

Apr 28, 2020

This is the final chapter. It is one of the easiest and it was fun doing that lunar landing project. This specialisation is the best for a person taking baby steps in the reinforcement learning.

By Chad R

Feb 27, 2020

Great course for learning the fundamentals. I liked that it tied into function approximation for deep reinforcement learning. The text book made the fundamental concepts more clear.

By Simon L

Jul 9, 2020

Loved this entire course. Learned a lot. In a relatively easy way.

Lots of fun of as well. Indeed, I have already used some of these new skills in the real world.

Couldn't be better!

By Durgesh P

Mar 26, 2024

After taking this course , I am able to understand maximum paper based on ml and also able to implement advanced algorithms and also able to implement real life problems using RL.

By Jo K

Jun 17, 2021

Very good course and specialization. If you want to get the most out of it, I recommend following their required reading and keep reading that book to cover other chapter as well.

By ding l

Jun 1, 2020

Matha and Adam, thank you again. I will try to apply what I learned here to my own work, a content recommendation system based on deep learning and reinforcement learning.