Deep Reinforcement Learning: DQN

In our previous post about Reinforcement Learning, we made an introduction to this area through one of its most popular techniques: Q-learning. We laid the groundwork by talking about Markov decision processes, policies and value functions, and we saw a…