CRAAM
2.0.0
Robust and Approximate Markov Decision Processes
|
A random policy with state-dependent action sets which are discrete. More...
#include <Simulation.hpp>
Public Types | |
using | State = typename Sim::State |
using | Action = typename Sim::Action |
Public Member Functions | |
RandomPolicy (const Sim &sim, random_device::result_type seed=random_device{}()) | |
Action | operator() (State state) |
Returns a random action. | |
A random policy with state-dependent action sets which are discrete.
Important: Retains the reference to the simulator from which it comes. If the simulator is destroyed in the meantime, the behavior is not defined.
Sim | Simulator class for which the policy is to be constructed. Must implement an instance method actions(State). |