This chapter establishes the groundwork for understanding Reinforcement Learning (RL). We start by defining RL and comparing it to other machine learning approaches. You will be introduced to the core components involved: the agent making decisions, the environment it operates within, the states describing the situation, the actions available to the agent, and the rewards received as feedback.

We will examine how an agent's behavior is determined by a policy and how the learning process unfolds through an interaction loop. The distinction between episodic and continuing tasks will also be explained. Finally, we cover the practical steps for setting up a Python environment using libraries like Gymnasium and NumPy. Grasping these concepts is the first step toward building agents that learn optimal behaviors.