public ExpectedSarsa(DiscreteEnvironment environment, ActionPolicy policy, double stepSize, double discount) { this.environment = environment; this.policy = policy; this.stepSize = stepSize; this.discount = discount; ActionValueFunction = new DiscreteActionValueFunction(environment); }
public DiscreteActionValueFunction(DiscreteEnvironment environment, double initValue = 0.0) { ActionValues = new List <Pair <State, List <Pair <AIAction, double> > > >(); this.environment = environment; this.InitValue = initValue; }
private DiscreteActionValueFunction(DiscreteEnvironment environment, List <Pair <State, List <Pair <AIAction, double> > > > actionValues) { this.ActionValues = actionValues; this.environment = environment; }
public DiscreteActionValueFunction(DiscreteEnvironment environment, double initValue = 0.0) { this.environment = environment; this.InitValue = initValue; }