public QLearning(IQLearningModel Model, List <IQState> States, List <IQAction> Actions, Decimal Gamma)
: base(Model, States, Actions, Gamma)
{
foreach (IQState state in States)
{
foreach (IQAction action in Actions)
{
if (!this.NumQ.ContainsKey(state))
{
this.NumQ.Add(state, new Dictionary <IQAction, int>());
}
this.NumQ[state].Add(action, 0);
}
}
}
public QLearningBase(IQLearningModel Model, List <IQState> States, List <IQAction> Actions, Decimal Gamma)
{
this.Model = Model;
this.States = States;
this.Actions = Actions;
this.Gamma = Gamma;
foreach (IQState state in States)
{
foreach (IQAction action in Actions)
{
if (!this.QValues.ContainsKey(state))
{
this.QValues.Add(state, new Dictionary <IQAction, decimal>());
}
this.QValues[state].Add(action, this.InitialQ);
}
}
}
public DelayedQLearning(IQLearningModel Model, List <IQState> States, List <IQAction> Actions, Decimal Gamma, Decimal Epsilon, Decimal Delta)
: base(Model, States, Actions, Gamma)
{
this.Epsilon = Epsilon;
this.Delta = Delta;
foreach (IQState state in States)
{
foreach (IQAction action in Actions)
{
if (!this.AttemptedUpdates.ContainsKey(state))
{
this.AttemptedUpdates.Add(state, new Dictionary <IQAction, decimal>());
}
this.AttemptedUpdates[state].Add(action, 0.0M);
if (!this.NumSamplesInNextUpdate.ContainsKey(state))
{
this.NumSamplesInNextUpdate.Add(state, new Dictionary <IQAction, int>());
}
this.NumSamplesInNextUpdate[state].Add(action, 0);
if (!this.TimeOfLastAttemptedUpdate.ContainsKey(state))
{
this.TimeOfLastAttemptedUpdate.Add(state, new Dictionary <IQAction, int>());
}
this.TimeOfLastAttemptedUpdate[state].Add(action, 0);
if (!this.LearnFlags.ContainsKey(state))
{
this.LearnFlags.Add(state, new Dictionary <IQAction, bool>());
}
this.LearnFlags[state].Add(action, true);
}
}
decimal epsilon1 = GetEpsilon1(Epsilon, Gamma);;
this.Epsilon1 = epsilon1;
this.NumSamplesNeeded = GetNumSamplesNeeded(Gamma, epsilon1, States.Count, Actions.Count, Delta);
}
