/the-elements-of-reinforcement-learning

Reinforcement Learning (RL) is believe to be a more general approach towards Artificial Intelligence (AI). RL is the foundation for many recent AI applications, e.g., Automated Driving, Automated Trading, Robotics, Gaming, Dynamic Decision, etc. With concrete examples, this repository tries introduce clearly the basic elements of Reinforcement Learning, e.g., Agent, Environment, State, State Transition, Policy, Action, Reward, Future Return, Discounted Future Return, Exploration & Exploitation, Markov Decision Processing, The Bellman Equation, Policy-based Learning, Value-based Learning, etc.

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

the-elements-of-reinforcement-learning

Reinforcement Learning (RL) is believe to be a more general approach towards Artificial Intelligence (AI). RL is the foundation for many recent AI applications, e.g., Automated Driving, Automated Trading, Robotics, Gaming, Dynamic Decision, etc.

With concrete examples, this repository tries introduce clearly the basic elements of RL, e.g., Agent, Environment, State, State Transition, Policy, Action, Reward, Future Return, Discounted Future Return, Exploration & Exploitation, Markov Decision Processing, The Bellman Equation, Policy-based Learning, Value-based Learning, etc.

reference

  • Reinforcement Learning: an introduction, Richard S. Sutton and Andrew G. Barto, 2018, pdf
  • Reinforcement Learning: an introduction, Richard S. Sutton and Andrew G. Barto, 2018, code
  • Deep Reinforcement Learning: An Overview, Yuxi Li, 2017, pdf
  • CS 294: Deep Reinforcement Learning, Fall 2017, UC Berkeley course

requirements:

Install python packages using this command:

pip install numpy scipy pandas matplotlib sklearn pillow