Markov Decision Processses (MDP) Toolbox

• Markov Decision Processses (MDP) Toolbox
  • mdp_Q_learning — Solves discounted MDP with the Q-learning algorithm (Reinforcement learning).
  • mdp_bellman_operator — Applies the Bellman operator to a value function Vprev and returns a new value function and a Vprev-improving policy.
  • mdp_check — Checks the validity of a MDP.
  • mdp_check_square_stochas — Checks whether a matrix is square and stochastic.
  • mdp_computePR — Computes the reward associated to a state/action pair.
  • mdp_computePpolicyPRpoli — Computes the transition matrix and the reward matrix for a given policy.
  • mdp_eval_policy_TD_0 — Evaluates a policy using the TD(0) algorithm.
  • mdp_eval_policy_iterative — Evaluates a policy using iterations of the Bellman operator.
  • mdp_eval_policy_matrix — Evaluates a policy using matrix operation.
  • mdp_eval_policy_optimali — Determines sets of 'near optimal' actions for all states.
  • mdp_example_forest — Generates a simple MDP example of forest management problem.
  • mdp_example_rand — Generates a random MDP problem.
  • mdp_finite_horizon — Solves finite-horizon MDP with backwards induction algorithm.
  • mdp_policy_iteration — Solves discounted MDP with policy iteration algorithm.
  • mdp_policy_iteration_mod — Solves discounted MDP with modified policy iteration algorithm.
  • mdp_relative_value_itera — Solves MDP with average reward with relative value iteration algorithm.
  • mdp_span — Computes the span of a vector.
  • mdp_value_iteration — Solves discounted MDP with value iteration algorithm.
  • mdp_value_iterationGS — Solves discounted MDP with Gauss-Seidel's value iteration algorithm.
  • mdp_value_iteration_boun — Computes a bound on the number of iterations for the value iteration algorithm.
  • mdp_verbose_silent — Defines verbose or silent running mode.