Skip to main content

Knowledge check

Quiz
In Progress

Knowledge check

Evaluate your understanding of this unit by completing the knowledge check below.

Knowledge check illustration

Action item: Knowledge check

Work through each question. Correct answers and feedback mirror the Multiverse assessment.

Question 1 of 10
1. Which statement best describes the goal of model training in machine learning?
  • A. To memorise all patterns in the training data for maximum accuracy.
  • B. To identify and store all training examples for use in predictions.
  • C. To adjust model parameters to minimize the loss function and improve predictions.
  • D. To reduce computational costs by skipping gradient updates.
Correct Answer: C

The best choice is (C): To adjust model parameters to minimize the loss function and improve predictions.

2. In gradient descent, what does the learning rate control?
  • A. The direction in which the gradient points.
  • B. The number of iterations required to reach convergence.
  • C. The size of the step taken toward the minimum of the loss function.
  • D. The total number of features used in training.
Correct Answer: C

The best choice is (C): The size of the step taken toward the minimum of the loss function.

3. Which of the following best describes a limitation of vanilla gradient descent?
  • A. It updates weights too frequently for large datasets.
  • B. It struggles with slow convergence and can get stuck in local minima.
  • C. It automatically adjusts learning rates for each parameter.
  • D. It is only effective for non-linear models.
Correct Answer: B

The best choice is (B): It struggles with slow convergence and can get stuck in local minima.

4. Which optimizer combines the advantages of momentum and RMSprop?
  • A. Adagrad
  • B. SGD with Nesterov acceleration.
  • C. Adadelta
  • D. Adam (Adaptive Moment Estimation)
Correct Answer: D

The best choice is (D): Adam (Adaptive Moment Estimation)

5. Which situation would most likely benefit from using RMSprop?
  • A. A large, noisy dataset where gradients vary significantly between batches.
  • B. A small, clean dataset with stable gradients.
  • C. A dataset with sparse features and rare signals.
  • D. A small dataset requiring strong feature selection.
Correct Answer: A

The best choice is (A): A large, noisy dataset where gradients vary significantly between batches.

6. What is the main purpose of L1 regularisation (Lasso regression)?
  • A. To reduce the influence of correlated features without removing them.
  • B. To eliminate irrelevant features by forcing their coefficients to zero.
  • C. To maintain all features but shrink their coefficients.
  • D. To reduce computation time during gradient updates.
Correct Answer: B

The best choice is (B): To eliminate irrelevant features by forcing their coefficients to zero.

7. What happens when the regularisation strength (λ) is set too high?
  • A. The model converges faster to the global minimum.
  • B. The model perfectly fits both training and test data.
  • C. The model ignores the bias term.
  • D. The model may underfit the data.
Correct Answer: D

The best choice is (D): The model may underfit the data.

8. Which statement best describes dropout in neural networks?
  • A. It permanently removes neurons from the network to simplify architecture.
  • B. It increases the learning rate for underperforming neurons.
  • C. It randomly deactivates neurons during training to prevent over-reliance on specific pathways.
  • D. It duplicates neurons to create redundant layers.
Correct Answer: C

The best choice is (C): It randomly deactivates neurons during training to prevent over-reliance on specific pathways.

9. Which regularisation technique is most useful when your dataset has many correlated features?
  • A. L2 (Ridge)
  • B. L1 (Lasso)
  • C. Early stopping
  • D. Dropout
Correct Answer: A

The best choice is (A): L2 (Ridge)

10. If your training loss is much lower than your validation loss, what’s the most likely cause?
  • A. Underfitting — the model hasn’t learned enough from the training data.
  • B. Data leakage — the test data was used during training.
  • C. Overfitting — the model has learned noise from the training data.
  • D. Poor optimiser selection — the model didn’t converge.
Correct Answer: C

The best choice is (C): Overfitting — the model has learned noise from the training data.