# Roadmap and Resources
For a fresher, to break into the field of reinforcement learning, the following resources are suggested, 

1. <strong><em>Books</em></strong>
    1. [Reinforcement Learning: An Introduction](https://web.stanford.edu/class/psych209/Readings/SuttonBartoIPRLBook2ndEd.pdf) By Richard S. Sutton and Andrew G. Barto
    2. [Deep Reinforcement Learning Hands-On](https://www.packtpub.com/product/deep-reinforcement-learning-hands-on/9781788834247) By Maxim Lapan
    3. [Deep Reinforcement Learning](https://www.springer.com/gp/book/9789811540943) Fundamentals, Research and Applications, Springer
    4. [Deep Reinforcement Learning](https://www.springer.com/gp/book/9789811382840) Frontiers of Artificial Intelligence, Springer
2. <strong><em>Python Libraries</em></strong>
    1. OpenAI ([Gym](https://gym.openai.com/docs/), [Spinning Up](https://spinningup.openai.com/en/latest/user/introduction.html))
    2. [stable-baselines3](https://stable-baselines3.readthedocs.io/en/master/guide/install.html)
    3. [RLlib](https://docs.ray.io/en/latest/rllib.html), and more 
3. <strong><em>Blogs and Courses</em></strong>
    1. [This](https://simoninithomas.github.io/deep-rl-course/) is a good introductory blog that can serve as an excellent first-read for freshers breaking into this field.
    2. [This](https://www.davidsilver.uk/teaching/) is an excellent course by David Silver.
    3. [This](https://www.coursera.org/specializations/reinforcement-learning) is a good Coursera specialization for getting started with RL.
    4. [This](https://youtube.com/playlist?list=PLqYmG7hTraZDVH599EItlEWsUOsJbAodm) is DeepMind x UCL RL 2021 Lecture Series YouTube Playlist


```{note}
* Book #1 offers a well-formulated, comprehensive and detailed introduction to the building blocks of RL. It serves as an excellent first-read for those unfamiliar with RL, however the theory goes into a lot of depth which may be time-consuming. So, it is suggested to use this in tandem with the libraries and other resources.
* Books #2,3,4 offer a brief introduction to RL before exploring the field of <b>Deep Reinforcement Learning</b> (DRL).
* Book #2 has been built on <b>Pytorch</b>, while Books #3,4 develop on <b>Tensorflow</b>.
* Book #2 offers step by step tutorial on DRL, covering a wide variety of functionalities and examples.
* Book #3 discusses all the nitty-gritty details associated with DRL in a consise manner, which would otherwise require thorough sweeping of research papers. It is an excellent read to understand DRL and how it differs from traditional RL. It comes with a supporting github repo which provides sample code for the examples used.
* Book #4 is similar to #3 in its intend to discuss about DRL. However, the two differ in their approaches. The latter focuses on detailed disuccion on the nuances of DRL, the former however, focuses more on the code build up corresponding to the content flow.
<br><br><br>
* Though multiple libraries are available, I'll personally suggest using gym and stable-baselines as a first step.
<br><br><br>
* Within blogs and courses, #1 is highly recommended for a fresher as it disusses the basics with code implementations, and offers both blog and video format.
* #2 and #4 can offer a very comprehensive detailing of RL fundamentals.
* #3 serves as a good starting course that follows along the Barto and Sutton RL book.
```