public class TransitionProbabilityImpl extends TransitionProbability
TransitionProbability
.Modifier and Type | Field and Description |
---|---|
protected RandomOutcomeFunction<State,Action,double[]> |
randomOutcomeFunction |
Constructor and Description |
---|
TransitionProbabilityImpl(umontreal.ssj.probdistmulti.DiscreteDistributionIntMulti[] multiVariateDistributions,
RandomOutcomeFunction<State,Action,double[]> randomOutcomeFunction,
StateSpaceImpl[] stateSpace) |
TransitionProbabilityImpl(MultiINIDistribution[] multiVariateDistributions,
RandomOutcomeFunction<State,Action,double[]> randomOutcomeFunction,
StateSpaceImpl[] stateSpace) |
Modifier and Type | Method and Description |
---|---|
ArrayList<State> |
generateFinalStates(State initialState,
Action action)
This method constructs an
ArrayList<State> of states towards which the stochastic process may
transition in period t+1 if action is selected in initialState at period
t ; note that these states may not yet exist in the state space. |
ArrayList<State> |
getFinalStates(State initialState,
Action action)
This method retrieves an
ArrayList<State> of existing states towards which the stochastic process may
transition in period t+1 if action is selected in initialState at period
t . |
double |
getTransitionProbability(State initialState,
Action action,
State finalState)
This method returns the transition probability from
initialState to finalState when
action is selected. |
protected RandomOutcomeFunction<State,Action,double[]> randomOutcomeFunction
public TransitionProbabilityImpl(umontreal.ssj.probdistmulti.DiscreteDistributionIntMulti[] multiVariateDistributions, RandomOutcomeFunction<State,Action,double[]> randomOutcomeFunction, StateSpaceImpl[] stateSpace)
public TransitionProbabilityImpl(MultiINIDistribution[] multiVariateDistributions, RandomOutcomeFunction<State,Action,double[]> randomOutcomeFunction, StateSpaceImpl[] stateSpace)
public double getTransitionProbability(State initialState, Action action, State finalState)
TransitionProbability
initialState
to finalState
when
action
is selected.getTransitionProbability
in class TransitionProbability
initialState
- the initial state.action
- the action chosen.finalState
- the final state.initialState
to finalState
when
action
is selected.public ArrayList<State> generateFinalStates(State initialState, Action action)
TransitionProbability
ArrayList<State>
of states towards which the stochastic process may
transition in period t+1
if action
is selected in initialState
at period
t
; note that these states may not yet exist in the state space. This method is used by
forward recursion procedures.generateFinalStates
in class TransitionProbability
initialState
- the initial state of the stochastic process; note that we assume that
initialState
is associated with period t
.action
- the action selected at period t
.ArrayList<State>
of states towards which the stochastic process may
transition in period t+1
.public ArrayList<State> getFinalStates(State initialState, Action action)
TransitionProbability
ArrayList<State>
of existing states towards which the stochastic process may
transition in period t+1
if action
is selected in initialState
at period
t
. This method is used by backward recursion procedures.getFinalStates
in class TransitionProbability
initialState
- the initial state of the stochastic process; note that we assume that
initialState
is associated with period t
.action
- the action selected at period t
.ArrayList<State>
of states towards which the stochastic process may
transition in period t+1
.Copyright © 2017–2018. All rights reserved.