mdp-0.1.0.0: Tools for solving Markov Decision Processes.

Safe HaskellNone
LanguageHaskell98

Algorithms.MDP.Examples.Ex_3_1

Description

The problem described by Bertsekas p. 22.

Synopsis

Documentation

data State Source

There are two distinct states

Constructors

A 
B 

Instances

data Control Source

There are two distinct actions we can take in each state

Constructors

U1 
U2 

Instances

transition :: Control -> State -> State -> Double Source

The transition matrix

costs :: Control -> State -> Double Source

The costs associated with each state and action

alpha :: Double Source

The discount factor

states :: [State] Source

The available states

controls :: [Control] Source

The available actions

mdp :: MDP State Control Double Source

The MDP representing the problem.