View Source Rein.Agents.QLearning (rein v0.1.0)

Q-Learning implementation.

This implementation uses epsilon-greedy sampling for exploration, and doesn't contemplate any kind of target network.