Listen free for 30 days

Reinforcement Learning with Python: A Short Overview of Reinforcement Learning with Python

Narrated by: William Bahl
Length: 2 hrs and 32 mins
Categories: Non-fiction, Technology
5 out of 5 stars (38 ratings)

£7.99/month after 30 days. Cancel anytime

Summary

Reinforcement Learning with Python

Reinforcement learning is one of those data science fields which will most certainly shape the world. The changes are already visible since we have self-driving cars, robots and much more we used to see only in some futuristic movies. Reinforcement learning is a widely used machine learning technique, a computational approach when it comes to the different software agents which are trying to maximize the total amount of possible rewards they receive while interacting with some uncertain as well as very complex environments.

This book is divided into seven chapters in which you will get to reinforcement techniques and methodology better. The first chapters will introduce you to the main concept laying being reinforcement learning techniques. Further, you will see what the difference between reinforcement learning and other machine learning techniques is. The book also provides some of the basic solution methods when it comes to the Markov decision processes, dynamic programming, Monte Carlo methods and temporal difference learning.

What you will learn in this book:

  • Types of fundamental machine learning algorithms in comparison to reinforcement learning
  • Essentials of reinforcement learning process
  • Marko decision processes and basic parameters
  • How to integrate reinforcement learning algorithm using OpenAI Gym
  • How to integrate Monte Carlo methods for prediction
  • Monte Carlo tree search
  • Dynamic programming in Python for policy evaluation, policy iteration and value iteration
  • Temporal difference learning or TD
  • And much, much more....

Listen to this book now and learn more about reinforcement learning with Python!

©2017 Anthony Williams (P)2017 Anthony Williams

What members say

Average customer ratings

Overall

  • 5 out of 5 stars
  • 5 Stars
    38
  • 4 Stars
    0
  • 3 Stars
    0
  • 2 Stars
    0
  • 1 Stars
    0

Performance

  • 5 out of 5 stars
  • 5 Stars
    38
  • 4 Stars
    0
  • 3 Stars
    0
  • 2 Stars
    0
  • 1 Stars
    0

Story

  • 4.5 out of 5 stars
  • 5 Stars
    27
  • 4 Stars
    11
  • 3 Stars
    0
  • 2 Stars
    0
  • 1 Stars
    0
Sort by:
  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Clear concise with intuitive approach.

This is the kind of book that really help you build up a skill. It is simply brilliant.
Theory is precise and concisely explicited including equations and exercices.
The exercises are relevant and simple enough that all you have to care about is your understanding of the principles explained during the lectures.
In addition, there are even tips and comments on how to use this course, how to use an online course, and how to build complete machine learning background.
This is my first experience with Anthony Williams and even though, I did not follow other courses because they concern stuff that I already know, I would highly recommend them based on the content of this corse and the quality of the teaching.
I am even considering, going back to things I know just to revisit them in a pleasant, well-organised, concise way with a new take on it.
Thanks you for the effort you put in your courses.

23 of 23 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

ENJOYING

This is a thorough course on a complicated topic. I was not fully prepared when I started taking it and had to listen up on some prerequisite material before I could keep up. If you're willing to do that you'll enjoy the course.
I kept hoping for more sophisticated demonstrations of the algorithm in code, but I suppose that is out of scope for the course.

20 of 20 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

A good compound of practical and theoretical

You need to know calculus and probability, otherwise, it would not make sense for you. Perfect materials structure, like it.

20 of 20 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Understanding the whole idea of AI.

Best introductory course on Reinforcement Learning you could ever find here. Be warned though that without an advanced knowledge of probability you won't get the most out of this course. The derivation of Bellman equation that forms the basis of Reinforcement Learning is the key to understanding the whole idea of AI.

20 of 20 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Less spoon feeding

This course is not for those who watch videos and write on their CV "reinforcement learning expert". If you take the time to do the exercises, experiment, be curious and follow closely you'll learn a lot. Few words said, but the words are effectively chosen. Less spoon feeding and more guiding.

19 of 19 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    4 out of 5 stars

reinforcement learning made simple. Simple solid


reinforcement learning made simple. Simple solid math when needed, with good python code.
Solid introduction to reinforcement learning traditional strategies and modern deep reinforcement learning.
Definitively recommend.

19 of 19 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Bringing it all together

The instructor knows his subject well and has a nice way of bringing it all together. It is particularly useful if one has the right background in Deep Learning, Machine Learning, Python programming, and Mathematics.

18 of 18 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Very instructive and challenging.

There are some mathematical concepts that I have forgotten but I didn't mind looking for them elsewhere, I consider this part of the challenge.

What is sure about this course is that it needs efforts and serious work.
It is not for people who are looking for a quick hands on tutorial.
If you are looking to really delve into RL, this course is a good start.
Very good course

17 of 17 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Taking time to digest

This is a great course. I took it twice since the material isn't trivial and taking time to digest. I recommend to listen Anthony Williams book in parallel to get deeper understanding of the material in the course and after each part doing the exercises from the course.

16 of 16 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars

Very solid material.

The instructor delivers the knowledge promised in a methodical curriculum. I understand the basic reinforcement learning algorithms and how they work.

7 of 7 people found this review helpful

Sort by:
  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for moore
  • moore
  • 01-10-19

Definitely would recommend.

I have tried some other AI courses on Audible but they all gloss over the math and it makes it seem like they have a surface level understanding of what is actually going on. In this course it is clear The author Anthony Williams knows the math behind the algorithms and does a good job at setting up the building blocks so that by the time you see Q-Learning at the end you can see why it was important to first learn about multi armed bandit, MC, TD, etc. first. I'm also a huge fan of coding things by myself as I've done courses with Jupyter Notebooks with "fill in the blank" in the past and that can be harder to actually understand since you might not see how the code ties in. Definitely would recommend.

23 of 23 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for Stephanie
  • Stephanie
  • 01-10-19

For RL/DRL journey

The course is both detailed and comprehensive in its treatment of MDP and the solutions to it - Dynamic Programming, Monte Carlo, Marko decision processes and basic parameters, Temporal Difference, and Approximation Methods. The author progressively moves from easy to hard examples, and all Python code are thoroughly explained. I particularly like the fact that the provided code matches those in the explanation, and that the code actually WORKS! I'm taking Udacity's Deep Reinforcement Learning ND program, and I found that this course - particularly the MC and TD sections - complements the DRLND perfectly! Highly recommended to anybody starting on their RL/DRL journey.

23 of 23 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for Judy Harris
  • Judy Harris
  • 01-10-19

Makes it easy to understand.


Good examples and the material order is very good. I think, that sometimes, it is better to lead us to the final equation in some steps and not just write it. Makes it easy to understand.

22 of 22 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for fredric
  • fredric
  • 01-10-19

This is packed.

I'm a Mechanical Engineer with average python knowledge. Been interested in AI for a couple years and finally decided I'd take the leap. I did read a few books including Fundamentals of Deep-Learning. But this? This is packed. From topic to topic I could literally feel myself building up to something super useful. Can't stop now though. Had previously tried solving the MountainCar openAI gym problem and couldn't so that leads me to your next course obviously.
I'm enjoying this. I don't even feel like I'm in a course, feels like I'm having a friendly conversation and picking useful info while at it.

20 of 20 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for Marlene Neace
  • Marlene Neace
  • 01-10-19

Divided into 7 chapters

Very interesting seven chapter topics, explained in depth and brief. Such as the instructor says, working on the algorithms and theoretical concepts on your own is requiered to completely understand them.

20 of 20 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for martha blewis
  • martha blewis
  • 01-10-19

Studying of AI

I like the lecturer's personality and how he understands that in the studying of AI, you can't take shortcut, suits what me! Havent really tried all the courses yet, I'm gonna go to it soon, will update this review then.

19 of 19 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for gilbert
  • gilbert
  • 01-10-19

I motivated to continue learning

Very good course. it has the perfect equilibrium between the conceptual explanations and the code. In addition, its teaching methodology, always encouraging to code by ourselves is the type of methodology that I was looking for. I think it is still a introductory course, but I motivated to continue learning

18 of 18 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for Robert Gustin
  • Robert Gustin
  • 01-10-19

Topic of reinforcement learning.

This is probably the best course that I have found on the topic of reinforcement learning. I believe this is the next best thing to actually taking a specialised higher degree course in reinforcement learning.

16 of 16 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for minnie kenney
  • minnie kenney
  • 01-10-19

Very well explained

Particularly because you try to relate theoritical with practical code examples.
Because many courses, including the OMSCS Masters course I'm doing, have only theoritical aspect of RL, and expect students to implement practically all by themselves.
It's very hard to apply them for real world problems (like a Game)

16 of 16 people found this review helpful

  • Overall
    5 out of 5 stars
  • Performance
    5 out of 5 stars
  • Story
    5 out of 5 stars
Profile Image for larry
  • larry
  • 01-10-19

Get my mind turning


I've taken several of AW's courses already, and he has hit another one out of the park with this course. I'm coming from a background in neuroscience, so I am already acquainted with many of the ideas of reinforcement learning, but his style and choice of examples is doing wonders to get my mind turning again.

15 of 15 people found this review helpful