Hyperparameter Tuning Fundamentals

Submit

2. What is the relationship between batch size and gradient noise during training?

Larger batches produce noisier gradients

Smaller batches introduce more noise but may improve generalization

Batch size has no effect on gradient noise

Only batch normalization affects gradient noise

Submit

4. Which method combines multiple hyperparameter tuning strategies for efficiency?

Exhaustive search

Successive halving

Manual tuning

Single parameter sweep

5. True or False: Hyperparameter tuning should always use the test set for evaluation.

True

False

Submit

7. What is the purpose of a learning rate schedule?

To increase learning rate throughout training

To adjust learning rate dynamically to improve convergence

To keep learning rate constant

To eliminate the need for regularization

Submit

9. Which optimizer adapts learning rates per parameter based on historical gradients?

Stochastic gradient descent

Adam

Vanilla gradient descent

Nesterov momentum

10. True or False: Momentum in optimization accelerates convergence by accumulating gradients.

True

False

11. What is a hyperparameter in machine learning?

A parameter learned by the model during training

A configuration set before training that controls the learning process

The final weight in a neural network

A feature in the input dataset

12. What role does cross-validation play in hyperparameter tuning?

Reduces the size of training data

Provides robust performance estimates across multiple data splits

Eliminates the need for a validation set

Increases model bias

13. Which is a disadvantage of grid search compared to random search?

It is faster

It explores fewer dimensions efficiently

Computational cost grows exponentially with parameters

It requires fewer function evaluations

Submit

15. What does early stopping prevent?

Underfitting

Overfitting by halting training when validation performance plateaus

High bias

Feature scaling issues

16. True or False: Batch size does not affect model generalization or training speed.

True

False

17. Which regularization technique adds a penalty proportional to the square of weights?

L1 regularization

L2 regularization

Dropout

Batch normalization

Submit

19. What is the primary risk of setting the learning rate too high?

Model underfitting

Training becomes too slow

Loss may oscillate and fail to converge

Overfitting increases

20. Which technique evaluates all possible combinations of hyperparameters?

Random search

Grid search

Bayesian optimization

Early stopping

DataX Hyperparameter Tuning Basics Quiz

1. Dropout is a regularization technique that randomly _____ neurons during training.

2.

What first name or nickname would you like us to use?

2. What is the relationship between batch size and gradient noise during training?

3. In neural networks, the number of hidden units per layer is a _____ that affects model capacity.

4. Which method combines multiple hyperparameter tuning strategies for efficiency?

5. True or False: Hyperparameter tuning should always use the test set for evaluation.

6. Tree-based models use hyperparameters like max depth and min samples split to prevent ______.

7. What is the purpose of a learning rate schedule?

8. The _____ hyperparameter controls the trade-off between bias and variance in regularization.

9. Which optimizer adapts learning rates per parameter based on historical gradients?

10. True or False: Momentum in optimization accelerates convergence by accumulating gradients.

11. What is a hyperparameter in machine learning?

12. What role does cross-validation play in hyperparameter tuning?

13. Which is a disadvantage of grid search compared to random search?

14. Bayesian optimization uses a _____ to model the objective function.

15. What does early stopping prevent?

16. True or False: Batch size does not affect model generalization or training speed.

17. Which regularization technique adds a penalty proportional to the square of weights?

18. Random search samples hyperparameters from a _____ distribution.

19. What is the primary risk of setting the learning rate too high?

20. Which technique evaluates all possible combinations of hyperparameters?