def reset(self): # TODO: Optimistic Initialization? self.weights = np.ones(self.num_features * len(self.actions)) QLearningAgent.reset(self)
def reset(self): self.weights = np.zeros(self.num_features * len(self.actions)) QLearningAgent.reset(self)
def reset(self): self.weights = np.zeros(self.num_features*len(self.actions)) QLearningAgent.reset(self)