Reinforcement Learning Applications Quiz

1. The discount factor (gamma) in RL controls:

How much future rewards are weighted relative to immediate rewards

The learning rate of the agent

The batch size during training

The exploration probability

2. Multi-agent reinforcement learning is most relevant for:

Robotics swarms and competitive game environments

Single-player puzzle solving only

Text classification tasks

Regression problems exclusively

3. The off-policy learning paradigm allows an agent to:

Learn from data generated by a different policy

Ignore past experiences

Train without a reward signal

Use only online data

4. Reward shaping in RL refers to:

Designing intermediate rewards to guide learning without changing optimal policy

Removing negative rewards entirely

Scaling rewards to unit variance

Using only sparse terminal rewards

5. In game-playing RL (e.g., AlphaGo), Monte Carlo Tree Search is used to:

Simulate future game states and evaluate move quality

Shuffle the training data

Compress the policy network

Generate random board positions

6. A Markov Decision Process (MDP) requires that the next state depends only on:

The current state and action

All previous states and actions

The reward function exclusively

The exploration strategy

7. Proximal Policy Optimization (PPO) improves upon policy gradients by:

Clipping the objective to prevent overly large updates

Removing the discount factor

Using only random actions

Eliminating the value function

8. The advantage function in policy gradient methods estimates:

How much better an action is compared to the average action in a state

The total cumulative reward

The probability of success

The exploration bonus

9. In finance, RL is applied to:

Portfolio optimization and algorithmic trading strategies

Creating new currencies

Auditing financial statements

Calculating simple interest rates

10. What is experience replay used for in DQN?

Breaking correlations between consecutive samples and improving stability

Storing all past rewards permanently

Increasing the exploration rate automatically

Normalizing reward values

11. In reinforcement learning, what is the primary role of the reward signal?

To guide the agent toward desirable behaviors and outcomes

To punish the agent for incorrect actions only

To provide labeled data for supervised learning

To initialize the agent's neural network weights

12. Deep Q-Networks (DQN) combine Q-learning with:

Deep neural networks for function approximation

Supervised learning labels

Recurrent neural networks exclusively

Convolutional filters only

13. Which RL application is most commonly used in autonomous vehicles?

Path planning and decision-making in traffic

Image classification only

Natural language processing

Time series forecasting

14. In actor-critic methods, the critic learns to estimate:

The value function or state values

The optimal policy directly

The reward function parameters

The exploration strategy

15. Policy gradient methods directly optimize:

The policy parameters to maximize expected reward

The reward function coefficients

The exploration rate only

The discount factor dynamically

16. What is the Bellman equation used for in RL?

Computing the optimal value function recursively

Normalizing input features

Calculating the gradient descent step size

Encoding categorical variables

17. Which algorithm is model-free and learns directly from experience?

Q-learning

Dynamic programming

Monte Carlo tree search

Markov chain analysis

18. In Q-learning, the Q-value represents:

The expected long-term reward for taking an action in a state

The quantity of training data needed

The query complexity of the algorithm

The quality score of the reward function

19. What is a policy in reinforcement learning?

A mapping from states to actions that an agent follows

A set of rules that define reward values

A loss function used to train neural networks

A database of past experiences

20. Which of the following best describes the exploration-exploitation tradeoff?

Choosing between known high-value actions and discovering potentially better ones

Deciding whether to use deep learning or traditional ML

Balancing training time and model complexity

Selecting between supervised and unsupervised learning

DataAI Reinforcement Learning Applications Quiz