pomdp (version 0.99.0)

policy: Extract the Policy from a POMDP/MDP

Description

Extracts the policy from a solves POMDP/MDP.

Usage

policy(x)

Arguments

x

A solved POMDP object.

Value

A list with the policy for each epoch. returned

comp2

Description of 'comp2'

%% ...

Details

A list (one entry per epoch) with the optimal policy. The policy is a data frame consisting of three parts.

Part 1: The value function with one column per state. (For MDPs this is just one column with the state).

Part 2: One column with the optimal action.

Part 3: One column per observation with the index of the row representing the policy node in the next epoch.

Examples

# NOT RUN {
data("Tiger")
sol <- solve_POMDP(model = Tiger)
sol

# policy with value function, optimal action and transitions for observations.
policy(sol)

# }