An agent is the learner or decision-maker in RL. It could be a robot, a program, or even a player in a game.

The environment is everything the agent interacts with – the world or setting of the problem.

For example, the environment could be a maze, a game board, or any situation where the agent acts.

At each moment, the agent is in some state, which describes its current situation (like a location on a map). The agent then chooses an action (a move or decision).

After the action, the environment gives the agent a reward – feedback that can be positive (gain points) or negative (lose points).

Key terms in reinforcement learning include:

• Agent: The decision-maker or learner that takes actions.

• Environment: The world or system where the agent operates.

• State: The situation or condition the agent is currently in.

• Action: A possible move or decision the agent can make.

• Reward: Feedback from the environment (like points gained or lost) based on the agent’s action.

These basic ideas match common RL descriptions.

In every step of learning, the agent sees its state, chooses an action, and then the environment updates the state and gives a reward. By repeating this loop, the agent learns which actions tend to give higher rewards.