private MDPUtilityFunction <TState> UpdateUtilityFunction(double gamma)
{
MDPUtilityFunction <TState> uf = utilityFunction.Copy();
double u_s = utilityFunction.GetUtility(PreviousState);
double gammaUtilDIff = ((gamma * utilityFunction
.GetUtility(CurrentState)) - utilityFunction
.GetUtility(PreviousState));
double alphaTerm = stateCount.ProbabilityOf(PreviousState)
* (previousReward + gammaUtilDIff);
uf.SetUtility(PreviousState, u_s + alphaTerm);
return(uf);
}
