Machine learning with rewards (as opposed to supervised or unsupervised methods).

Page List

• Definitions & Concepts — Concise list of terms and concepts in reinforcement learning.
• Dynamic Programming — Markov decision processes formally describe an environment for Reinforcement Learning when the environment is fully observable.
• Fundamentals — Introduction to the fundamentals of reinforcement learning.
• Markov Decision Processes — Markov decision processes formally describe an environment for Reinforcement Learning when the environment is fully observable.
• Model Free — Markov decision processes formally describe an environment for Reinforcement Learning when the environment is not observable.
• Value Function Approximation — Approximating the state or action value function.

Resources

• David Silver's Course Slides