rlai | This is a Python implementation of concepts and algorithms described in “Reinforcement Learning: An Introduction” (Sutton and Barto, 2018, 2nd edition).

Home > Chapter 3: Finite Markov Decision Processes

State of an MDP.

MDP environment.

Model-based MDP environment. Adds the specification of a probability distribution over next states and rewards.

Policy for use with tabular methods.

Gridworld MDP environment.