Regularization (L1 & L2) — Concept Check
Questions on Regularization (L1 & L2). Regularization adds a penalty for complexity to the loss so the model prefers simpler solutions — the main lever against overfitting, with L1 driving sparsity and L2 shrinking weights.
Questions in this quiz
8 questions on Regularization (L1 & L2) — Concept Check. Work through them here, then start the quiz to answer them and see your score.
1.A machine learning model is overfitting on the training data. Which regularization technique should you apply to encourage sparsity in the model?
- AL1 Regularization
- BIncrease learning rate
- CL2 Regularization
- DReduce dataset size
2.You have a model with weights [2.0, -3.0, 0.1, 0.0, 1.5]. What is the L1 penalty if λ=0.1?
- A0.85
- B0.56
- C0.46
- D0.35
3.A model trained without regularization is performing poorly on test data. What could be a likely reason?
- AThe model has overfit the data
- BThe model has a high variance
- CThe model has underfit the data
- DThe model has a high bias
4.Which statement about L2 regularization is true?
- AIt penalizes the sum of squared weights
- BIt forces many weights to exactly zero
- CIt increases model complexity
- DIt is primarily used for feature selection
5.A model with λ set to 10.0 is underperforming. What is a likely cause?
- AToo much regularization
- BInsufficient training data
- CIncorrect feature scaling
- DToo little regularization
6.What effect does increasing λ have on a model using L1 regularization?
- AIncreases sparsity of the model
- BHas no effect on the model
- CDecreases sparsity of the model
- DIncreases model complexity
7.A model with L2 regularization is not generalizing well. Which adjustment might help?
- ASwitch to L1 regularization
- BDecrease λ to reduce penalty
- CIncrease λ to increase penalty
- DAdd more layers to the model
8.Which regularization technique is known for automatic feature selection?
- AL2 Regularization
- BL1 Regularization
- CData Augmentation
- DDropout