mdp_value_of_policy: mdp_value_of_policy
In cboettig/mdplearning: Bayesian Learning Algorithms for Markov Decision Processes

Description Usage Arguments Details Value Examples

Compute the expected net present (e.g. discounted) value of a (not-necessarily optimal) policy in a perfectly observed (MDP) system

1 2	mdp_value_of_policy(policy, transition, reward, discount, model_prior = NULL, max_iter = 500, epsilon = 1e-05)

`policy`	the policy for which we want to determine the expected value
`transition`	list of transition matrices, one per model
`reward`	the utility matrix U(x,a) of being at state x and taking action a
`discount`	the discount factor (1 is no discounting)
`model_prior`	the prior belief over models, a numeric of length(transitions). Uniform by default
`max_iter`	maximum number of iterations to perform
`epsilon`	convergence tolerance

transition a list of transition matrices

the expected net present value of the given policy, for each state

source(system.file("examples/K_models.R", package="mdplearning"))
transition <- lapply(models, `[[`, "transition")
reward <- models[[1]][["reward"]]
df <- mdp_compute_policy(transition, reward, discount)
v <- mdp_value_of_policy(df$policy, transition, reward, discount)