Reinforcement learning
From Wikiversity
 Please help develop this page
This page was created, but so far, little content has been added. Everyone is invited to help expand and create educational content for Wikiversity. If you need help learning how to add content, see the editing tutorial and the MediaWiki syntax reference. To help you get started with content, we have automatically added references below to other Wikimedia Foundation projects. This will help you find materials such as information, media and quotations on which to base the development of "Reinforcement learning" as an educational resource. However, please do not simply copy-and-paste large chunks from other projects. You can also use the links in the blue box to help you classify this page by subject, educational level and resource type. |
|
Reinforcement learning
[edit] What is Reinforcement Learning
"Reinforcement learning (RL) is learning from interaction with an environment, from the consequences of action, rather than from explicit teaching." -- Rich Sutton
[edit] Evaluative Feedback (Chapter 2)
[edit] Softmax Action Selection
Softmax action selection is the way to maintain exploration and exploitation balance. The softmax policy will choose action a on period t with probablity: 
[edit] Reinforcement Learning Problems (Sutton and Barto Chapter 3)
[edit] Dynamic Programming (Sutton and Barto Chapter 4)
[edit] Monte Carlo Methods (Sutton and Barto Chaper 5)
[edit] Temporal-Difference Learning (Sutton and Barto Chapter 6)
[edit] Eligibility Traces (Sutton and Barto Chapter 7)
[edit] Related Terms
- Machine Learning
- Adaptive Dynamic Programming
- Markov Decision Process
[edit] References
- Sutton and Barto, Reinforcement Learning, an introduction, MIT Press 1998 (online version at http://www.cs.ualberta.ca/%7Esutton/book/ebook/the-book.html
