public override void PostSimulationStep()
{
// prepare for new inference of motivaiton from the hierarchy
base.PostSimulationStep();
base.m_motivatonSource.MakeStep();
m_prev_st = Ds.GetCurrentState();
this.m_newVariables.Clear();
}
public void Learn()
{
int[] s_tt = Ds.GetCurrentState();
float r_t = Ds.GetReward();
int a_t = Ds.GetLastExecutedAction();
if (a_t == this.m_prevSelectedAction)
{
base.SetJustExecuted(true);
}
if (a_t < 0)
{
// aciton ignored and nothing happened
if (r_t == 0)
{
return;
}
// action adding, reward received as a result of unknown action
List <int> toBeAdded = Ds.GetLastExecutedActions();
this.PerformActionAdding(toBeAdded);
return;
}
if (m_setup.OnlineSubspaceVariables)
{
m_mlvh.monitorVariableChanges(this);
}
LearningAlgorithm.Learn(r_t, s_tt, a_t);
if (r_t > 0)
{
// SRP is also an action, if the reward is received means that it just has been executed
base.SetJustExecuted(true);
this.ResetMotivationSource();
m_mlvh.performOnlineVariableRemoving(this);
}
}
public float[] SelectAction()
{
int[] s_tt = Ds.GetCurrentState();
float[] utils;
// use own asm to select one action?
if (m_setup.UseHierarchicalASM)
{
utils = this.Rescale(m_asm.SelectAction(Mem.ReadData(s_tt)));
}
else
{
utils = this.Rescale(Mem.ReadData(s_tt));
}
if (m_setup.PropagateUtilitiesInHierarchy)
{
float myMot = this.GetMyTotalMotivation();
Ds.PromoteUtilitiesToChilds(utils, myMot);
}
this.MarkSelection(utils);
return(utils);
}