RL

Reinforcement learning agents that learn from experience.

Agent	Description
DQN	Deep Q-Network with experience replay
PQN	Parallelized Q-Network (on-policy, no replay buffer)