Artificial Intelligence Reinforcement Learning 1

Question 1

What is the main objective of an agent in reinforcement learning?

Accepted Answer

To maximize cumulative reward over time through interaction with an environment

Answer

An RL agent learns a policy that maximizes the expected sum of discounted future rewards by taking actions and observing their outcomes in the environment.

Question 2

What is a 'policy' in reinforcement learning?

Accepted Answer

A mapping from states to actions that defines how the agent behaves

Answer

A policy specifies what action the agent takes in each state, either deterministically (one action per state) or stochastically (a probability distribution over actions).

Question 3

What does the 'reward signal' represent in reinforcement learning?

Accepted Answer

Feedback from the environment indicating the immediate value or desirability of the agent's last action

Answer

The reward signal provides scalar feedback after each action, guiding the agent toward behaviors that yield higher cumulative reward over time.

Question 4

What is the difference between model-based and model-free reinforcement learning?

Accepted Answer

Model-based RL learns or uses a model of the environment's dynamics to plan; model-free RL learns directly from experience without a model

Answer

Model-based RL builds a transition and reward model to simulate or plan ahead, while model-free RL learns value functions or policies directly from sampled interactions without an explicit environment model.

Question 5

What is an 'episode' in reinforcement learning?

Accepted Answer

A complete sequence of agent-environment interactions from the initial state to a terminal state

Answer

An episode is one full trajectory from start to termination (e.g., a game from start to win/lose), after which the environment resets for the next episode.

Artificial Intelligence Practice Test

Artificial Intelligence Practice Test

Artificial Intelligence Reinforcement Learning 1