Chevron Left
Voltar para Prediction and Control with Function Approximation

Comentários e feedback de alunos de Prediction and Control with Function Approximation da instituição Universidade de AlbertaUniversidade de Alberta

4.8
estrelas
726 classificações

Sobre o curso

In this course, you will learn how to solve problems with large, high-dimensional, and potentially infinite state spaces. You will see that estimating value functions can be cast as a supervised learning problem---function approximation---allowing you to build agents that carefully balance generalization and discrimination in order to maximize reward. We will begin this journey by investigating how our policy evaluation or prediction methods like Monte Carlo and TD can be extended to the function approximation setting. You will learn about feature construction techniques for RL, and representation learning via neural networks and backprop. We conclude this course with a deep-dive into policy gradient methods; a way to learn policies directly without learning a value function. In this course you will solve two continuous-state control tasks and investigate the benefits of policy gradient methods in a continuous-action environment. Prerequisites: This course strongly builds on the fundamentals of Courses 1 and 2, and learners should have completed these before starting this course. Learners should also be comfortable with probabilities & expectations, basic linear algebra, basic calculus, Python 3.0 (at least 1 year), and implementing algorithms from pseudocode. By the end of this course, you will be able to: -Understand how to use supervised learning approaches to approximate value functions -Understand objectives for prediction (value estimation) under function approximation -Implement TD with function approximation (state aggregation), on an environment with an infinite state space (continuous state space) -Understand fixed basis and neural network approaches to feature construction -Implement TD with neural network function approximation in a continuous state environment -Understand new difficulties in exploration when moving to function approximation -Contrast discounted problem formulations for control versus an average reward problem formulation -Implement expected Sarsa and Q-learning with function approximation on a continuous state control task -Understand objectives for directly estimating policies (policy gradient objectives) -Implement a policy gradient method (called Actor-Critic) on a discrete state environment...

Melhores avaliações

WP

11 de abr de 2020

Difficult but excellent and impressing. Human being is incredible creating such ideas. This course shows a way to the state when all such ingenious ideas will be created by self learning algorithms.

AC

1 de dez de 2019

Well peaced and thoughtfully explained course. Highly recommended for anyone willing to set solid grounding in Reinforcement Learning. Thank you Coursera and Univ. of Alberta for the masterclass.

Filtrar por:

101 — 125 de 133 Avaliações para o Prediction and Control with Function Approximation

por PHILIP C

18 de jun de 2021

T​his is a good course, but I continue to be disappointed in the lack of detail in the lectures. I fill in the detail with the Deep Mind lectures on Reinforcement Learning by David Silver. The programming assignments are difficult, not because they are challenging, but because the data structures are not well explained and the conceptual connections between the equations in the book and the code structures used for the implementation are not clear. It's like being given somebody's not-very-well-documented code and trying to figure out what they were thinking. All that said, I think that the course offers a lot and I have learned a lot from it so far.

por Luiz C

3 de out de 2019

Almost perfect, except two ~minor objections:

1/ the learning content between the 4 weeks is quite unbalanced. The initial weeks of the course are well sized, whereas week #3 and week #4 feel a touch light. It feels like the Instructors rushed to make the Course available online, and didn't have time to put as much content as they wished in the last weeks of the Course

2/ there are too many typos in some notebooks (specifically notebook of week #3). It gives the impression it was made in a rush, and nobody read over it again. Besides there seems to currently be some issue with this assignment

por Luka K

4 de jan de 2021

It is a good introduction to prediction and control with function approximation. Combining book and instructros results in a simple and nice explanation. What keeps it from the perfect grade are the examples. It would be nice if there are more examples and explained in a more detailed way why and how the example works. For example sometimes instructors would just say that the robot can use this, and that is mostly it. The other thing is more interactive project work. For example I would like to see how is my pendulum moving after N number of episodes. I would feel more satisfactory then.

por Dmitry S

5 de jan de 2020

Definitely a course to take to learn the ropes of RL. For this course, it is critical to follow and math. I'd love to give 5 stars to this course but will however take one away since the course could benefit a lot if the math was made a bit simpler to follow. The book referenced in the course is excellent and does help, but still, some more pedagogical repetition/rephrase, simplification of notation, a bit slower pace of narration would make the course even better. Having said that, this seems to be the best course available at this time. Many thanks to tutors.

por Hadrien H

4 de fev de 2021

I really appreciate that this course gives more hands on and assignments exercises. Really helped a lot in the understanding of the theory. As the books gets deeper into concepts and complexity so does the class, which is nice, but I felt like the depth and complexity in which the online class goes does not really keep up with the book content. Not only by skipping chapters but also by staying a bit at a too high level sometimes. Still a very good course again and really accessible, entertaining and resourceful material and instructors.

por Steven W

11 de mai de 2021

It's a great course, and they cover the basics of function approximation. The instructors were clear and knowledgeable, and the content that was covered was solid.

However, they skip some content that I feel is really important for modern RL, specifically the "deadly triad" regarding the convergence of off-policy approximate TD methods. They also don't discuss or link to papers on PPO or other recent advancements in RL, and I was hoping to learn more about those in particular.

por Narendra G

19 de jul de 2020

This course is important for those who not just want to learn RL for mere sake but want to dive into various topics currently in research (for that reading textbook is of most importance). This specialization would have been even better if it had included some more complex topics from the textbook. To fully comprehend all the topics, guidance from experts is necessary.

por Nicolas M

24 de out de 2020

Very interesting course: I have learned many things. A translation to other languages would be great: sometimes I can't memorize everything as I would if it was in my mother tongue.

Using another paper to study ( Experiments with Reinforcement Learningin Problems with Continuous State and Action Space) was a great idea that should be done in other courses.

por Lucas O S

21 de jan de 2020

Great course, deserve 5 stars. It is a good complement to the book, it adds interesting visualizations to help parse the content. The only issues were in the exercises. There are technical issues with the notebook platform where it keeps disconnecting from time to time, with no warning, and you lose your unsaved work (seems like token expiration).

por Anirban D

24 de jul de 2022

Excellent instructors and good concepts and assignments help you learn by doing. My only reason for giving a 4 is that this courses uses some internal tool (RL Glue) and hence none of the Jupyter notebooks are implementable outside. Some well known reinforcement learning framework like tensorforce perhaps should have been used.

por 남상혁

17 de jan de 2021

Very good lecture! I understand a lot about function approximation such as linear approximation, neural networks, etc. However, detail of video lectures were not perfect as the textbook. If you don't want to read a lot of text and listen to the lectures, you might not understand a lot of concepts.

por Hugo V

15 de jan de 2020

it was great to apply what I have learned from the book, but it was hard to find my mistakes in the course 3 notebook. I also misunderstood the alphas in the course 4 notebook at first glance, their indices look like they are powers (sorry for the bad english). Besides it, great course.

por Amit J

17 de mar de 2021

Lecture quality could have been better. They look like practiced monologues rather than a class where a teacher is trying (hard) to explain a concept. If one has to wait for assignment to get the full grasp, it doesn't reflect too well on the instructors.

por Lik M C

18 de jan de 2020

The course is still good. But the assignment is not as good as course 1 and 2. In fact, the contents of the course are getting complicated and interesting as well. But the assignments are relatively simple.

por Mark P

17 de ago de 2020

Solid intro course. Wish we covered more using neural nets. The neural net equations used very non-standard notation. Wish the assignments were a little more creative. Too much grid world.

por Anton P

12 de abr de 2020

There is a lot of material covered in the course. Be aware the pace picks up considerably from the first two courses. This said, it is a worthwhile course to take.

por Vladyslav Y

8 de set de 2020

I wish agents that are based on visual information (with the usage of CNN) would be included in the course. But overall that was really great!

por Sharang P

27 de fev de 2020

more detailed explanation of some of the assignments and how state values are got with tile coding but overall a great experience!

por Jerome b

9 de abr de 2020

Great course, based on the reference book about reinforcement learning. A must for anyone interested in machine learning.

por Rajesh M

17 de abr de 2020

I loved the course videos and programming assignments. The only suggestion would be to go a little deeper in the videos.

por SCOTT A

5 de ago de 2020

This was a good course but I really struggled to understand how each of the value functions translated into code.

por Muhammed A Ç

4 de set de 2021

Programming exercises are not self explaining. But instructors are explaining concept in a perfect way

por Pouya E

2 de dez de 2020

Great overall. The content on policy gradient could be expanded, some details were delivered hastily.

por Rishabh K

19 de mai de 2020

The average reward and differential return needs to be explained more thoroughly

por Ramaz J

17 de out de 2019

Course is great! Maybe some slides would be helpful not to forget.