What is Reinforcement Learning?

Reinforcement learning is concerned with how an agent ought to take actions in an environment so as to maximize its long-term reward. It is a very general framework, and as such, has interconnections with many fields, encompassing

  • computer science (artificial intelligence, machine learning),
  • engineering (information theory, adaptive control),
  • economics (rational agents, game theory),
  • mathematics (statistics),
  • psychology (behaviorism, motivation, incentives),
  • philosophy (inductive inference, theory of knowledge).

Reinforcement learning (augmented by ideas from information theory) can provide a unified theory for artificial intelligence, and is key to accomplishing more robust and flexible solutions, e.g. in computer games and autonomous robots.

Introductory Literature

  • R. Sutton and A. Barto. Reinforcement learning: An introduction
    Cambridge, MA, MIT Press (1998),
    Classic introductory textbook to RL. Highly Recommended.
    It requires no background knowledge, describes the key ideas, open problems, and great applications of this field.

A longer list of recent and classical papers on Reinforcement Learning is here

Other Resources

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License