What do Q-tables traditionally map in reinforcement learning?

In reinforcement learning, Q-tables are fundamental components used to represent the expected utility of taking certain actions in given states. Since the primary purpose of Q-tables is to facilitate decision-making in environments, they map states to actions, specifically the expected rewards for each action taken in a particular state.

Thus, Q-tables store values that indicate the quality (or Q-value) of taking a specific action from a specific state. This helps an agent learn the optimal strategy over time by exploring the environment and updating the Q-values based on the rewards received for actions taken. The correct mapping in the context of Q-tables is from states to actions, where each entry in the table represents the expected future rewards for taking a certain action in a certain state, guiding the agent toward maximizing its cumulative reward.

The idea behind reinforcement learning is that an agent learns to select the best actions based on its experiences, and the Q-table is a record of that learning process, making it an essential tool for evaluating actions within various states.