public class TheSpiderAndTheFly extends StochasticShortestPath<examples.jmdp.Distance,examples.jmdp.SpiderAction>
Constructor and Description |
---|
TheSpiderAndTheFly(int between,
double moveCost,
double stayCost,
double p) |
Modifier and Type | Method and Description |
---|---|
Actions<examples.jmdp.SpiderAction> |
feasibleActions(examples.jmdp.Distance i)
This method return all possible action which could be taken from a StateC
i, it is preferred that you create an action which makes the terminal
state becomes an absorbing state.
|
double |
immediateCost(examples.jmdp.Distance i,
examples.jmdp.SpiderAction a)
This method is easy to design, the immediate reward is directly
associated with the incurred action, in that way, you have to associate a
reward with the action taken.
|
static void |
main(java.lang.String[] args)
In this method you have to generate the problem then insert it into the
solver and finally solve the problem.
|
double |
prob(examples.jmdp.Distance i,
examples.jmdp.Distance j,
examples.jmdp.SpiderAction a)
This method return the associate transition probability between two
states by taking an action.
|
States<examples.jmdp.Distance> |
reachable(examples.jmdp.Distance i,
examples.jmdp.SpiderAction a)
Set of states that can be reached from this state i, after
taking the action a.
|
modifiedProb
getSteadyStateProbabilities, setProbabilitySolver, solve
getAllStates, getNumStates, getSolver
debug, debug, debug, getDebugLevel, getOptimalPolicy, getOptimalValueFunction, getReporter, isFinite, isSolved, operation, printSolution, printSolution, setDebugLevel, setReporter, setSolver, solve
public TheSpiderAndTheFly(int between, double moveCost, double stayCost, double p)
between
- moveCost
- p
- stayCost
- public double immediateCost(examples.jmdp.Distance i, examples.jmdp.SpiderAction a)
immediateCost
in class DTMDP<examples.jmdp.Distance,examples.jmdp.SpiderAction>
i
- Current Statea
- Current ActionDTMDP.immediateCost(State, Action)
public double prob(examples.jmdp.Distance i, examples.jmdp.Distance j, examples.jmdp.SpiderAction a)
prob
in class DTMDP<examples.jmdp.Distance,examples.jmdp.SpiderAction>
i
- Current state.j
- Destination Statea
- ActionDTMDP.prob(State, State, Action)
public States<examples.jmdp.Distance> reachable(examples.jmdp.Distance i, examples.jmdp.SpiderAction a)
DTMDP
public Actions<examples.jmdp.SpiderAction> feasibleActions(examples.jmdp.Distance i)
feasibleActions
in class InfiniteMDP<examples.jmdp.Distance,examples.jmdp.SpiderAction>
i
- Current StateInfiniteMDP.feasibleActions(State)
public static void main(java.lang.String[] args)
args
-