S - States class.A - Actions class.public class PolicyIterationSolver<S extends State,A extends Action> extends AbstractDiscountedSolver<S,A>
| Constructor and Description |
|---|
PolicyIterationSolver(DTMDP<S,A> problem,
double discountFactor)
The constructor method exclusively receives a problem of the
type InfiniteMDP because this solver is only designed to work
on infinite horizon problems.
|
PolicyIterationSolver(DTMDP<S,A> problem,
double discountFactor,
boolean setModifiedPolicy)
The constructor method exclusively receives a problem of the
type InfiniteMDP because this solver is only designed to work
on infinite horizon problems.
|
| Modifier and Type | Method and Description |
|---|---|
java.lang.String |
description()
This method return a complete verbal describtion of this element.
|
double |
getIncreasingFactor() |
double |
getInitialIterations() |
long |
getIterations() |
long |
getProcessTime() |
java.lang.String |
label()
The sub classes must return the Solver name.
|
void |
setIncreasingFactor(double increasingFactor)
Sets the increasing factor of the maximum iterations of the
Modified policy iteration method.
|
void |
setInitialIterations(int initialIterations)
Sets maximum iterations for the first run of the modified
policy iteration.
|
void |
setModifiedPolicy(boolean val)
Activates the modified policy iteration algorithm.
|
Solution<S,A> |
solve()
Called to solve the problem.
|
getInterestRate, setInterestRategetProblem, printSolutiongetOptimalPolicy, getOptimalValueFunction, getValueFunction, isSolved, printSolution, setPrintProcessTime, setPrintValueFunction, toStringequals, getClass, hashCode, notify, notifyAll, wait, wait, waitequalspublic PolicyIterationSolver(DTMDP<S,A> problem, double discountFactor)
problem - the structure of the problem of type InfiniteMDPdiscountFactor - represents how much less is the reward
received in the next period instead of receiving it in
the present period.public PolicyIterationSolver(DTMDP<S,A> problem, double discountFactor, boolean setModifiedPolicy)
problem - the structure of the problem of type InfiniteMDPdiscountFactor - represents how much less is the reward
received in the next period instead of receiving it in
the present period.setModifiedPolicy - public double getIncreasingFactor()
public void setIncreasingFactor(double increasingFactor)
increasingFactor - greater that 1. Determines max
iterations growth.public double getInitialIterations()
public void setInitialIterations(int initialIterations)
initialIterations - public Solution<S,A> solve() throws SolverException
Solverpublic void setModifiedPolicy(boolean val)
val - True if the modified policy iteration is to be used.public java.lang.String description()
JMarkovElementdescription in interface JMarkovElementdescription in class Solver<S extends State,A extends Action>JMarkovElement.label()public java.lang.String label()
Solverlabel in interface JMarkovElementlabel in class Solver<S extends State,A extends Action>Solver.toString()public final long getProcessTime()
public final long getIterations()
getIterations in class AbstractInfiniteSolver<S extends State,A extends Action>