Machine Learning Reinforcement Learning 2

Question 1

What is Q-learning in reinforcement learning?

Accepted Answer

A model-free RL algorithm that learns action-value (Q) functions

Answer

Q-learning is a model-free RL algorithm that learns the value of taking action a in state s (Q-values) to find an optimal policy without requiring a model of the environment.

Question 2

What does the Bellman equation capture in reinforcement learning?

Accepted Answer

The recursive relationship between the value of a state and its successor states

Answer

The Bellman equation expresses the value of a state as the immediate reward plus the discounted value of the next state, establishing recursive relationships used to solve RL problems.

Question 3

What is temporal difference (TD) learning?

Accepted Answer

Learning from differences between consecutive time steps using bootstrapping before episode completion

Answer

TD learning updates value estimates based on the difference between predicted and observed rewards at consecutive time steps, combining ideas from Monte Carlo methods and dynamic programming.

Question 4

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

Accepted Answer

Model-based learns an environment model to plan ahead; model-free learns directly from interaction

Answer

Model-based RL learns or uses a model of the environment's dynamics to plan ahead, while model-free RL learns policies or value functions directly from experience without building a model.

Question 5

In the SARSA algorithm, what does each letter represent?

Accepted Answer

State, Action, Reward, State, Action

Answer

SARSA stands for State, Action, Reward, State, Action — representing the tuple (s, a, r, s', a') used in its on-policy TD update rule.

Machine Learning Practice Test

Machine Learning Practice Test

Machine Learning Reinforcement Learning 2