public static class TimeIndexedTDLambda.StateTimeElibilityTrace extends TDLambda.StateEligibilityTrace
TDLambda.StateEligibilityTrace
to include time/depth information.Modifier and Type | Field and Description |
---|---|
int |
timeIndex
The time/depth of the state this eligibility represents.
|
eligibility, sh, v
Constructor and Description |
---|
StateTimeElibilityTrace(HashableState sh,
int time,
double eligibility,
burlap.behavior.singleagent.learning.actorcritic.critics.TDLambda.VValue v)
Initializes with hashed state, eligibility value, time/depth of the state, and the value function value associated with the state.
|
public int timeIndex
public StateTimeElibilityTrace(HashableState sh, int time, double eligibility, burlap.behavior.singleagent.learning.actorcritic.critics.TDLambda.VValue v)
sh
- the hashed input state for this eligibilitytime
- the time/depth of the state at which it was first visited.eligibility
- the eligibility of the statev
- the value function value for the state.