What is the primary goal of Reinforcement Learning?

• To classify data into categories

• To maximize cumulative rewards

• To reduce the error of predictions

• To cluster similar items together

In Reinforcement Learning, what is the term for the feedback received from the environment after an action is taken?

• Observation

• Reward

• State

• Action

Which of the following algorithms is a model-free approach in Reinforcement Learning?

• Markov Decision Process (MDP)

• Q-Learning

• Hidden Markov Model (HMM)

• Neural Network

What does the term "exploration" refer to in the context of Reinforcement Learning?

• The process of optimizing known actions

• Trying new actions to discover their effects

• Analyzing the results of past actions

• Storing the best actions for future use

What is the function of the "critic" in Actor-Critic methods?

• To propose new actions

• To assess the actions taken by the actor and provide feedback

• To explore the environment

• To create a model of the environment

Which of the following is NOT a common application area for Reinforcement Learning?

• Autonomous Robotics

• Image Classification

• Game Playing

• Personalized Healthcare

What is a potential consequence of using reinforcement learning in healthcare?

• Increased accuracy in diagnosis

• Improved patient satisfaction

• Ethical dilemmas due to exploration of experimental treatments

• Reduced healthcare costs

Which algorithm combines deep learning with Q-learning to handle high-dimensional state spaces?

• Trust Region Policy Optimization (TRPO)

• Deep Q-Networks (DQN)

• Proximal Policy Optimization (PPO)

• SARSA

In which of the following scenarios could you apply Reinforcement Learning?

• Predicting customer churn based on historical data

• Optimizing the inventory management strategy for a retail store

• Training a robotic arm to perform tasks by learning from trial and error

• Classifying email as spam or not spam

What dilemma does an RL agent face when designing treatment plans for patients?

• Balancing speed and accuracy

• Dawdling between simple and complex models

• The exploration-exploitation trade-off

• Deciding between supervised and unsupervised learning techniques