Describe fundamental principles of machine learning on Azure →mediumMultiple ChoiceObjective-mapped

AI-900 Practice Question: Describe fundamental principles of machine learning on Azure

What is 'regularisation' in machine learning and what problem does it solve?

⚠ Common exam trap

Candidates often confuse regularisation with data preprocessing steps like normalisation or reshaping, because both involve modifying data or model parameters, but regularisation specifically targets overfitting by penalising complexity, not by altering input data format or scale.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

Adding a penalty to the loss function to discourage overly complex models and reduce overfitting

Regularisation is a technique used in machine learning to prevent overfitting by adding a penalty term to the loss function. This penalty discourages the model from learning overly complex patterns, such as large or numerous coefficients, which helps the model generalise better to unseen data. In Azure Machine Learning, regularisation parameters like L1 (Lasso) or L2 (Ridge) can be configured in algorithms such as linear regression or neural networks to control model complexity.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✗
Standardising input features to the same scale before training
Why it's wrong here
Standardising input features to the same scale before training is a preprocessing step, not a regularisation technique. Rescaling features (via z-score normalisation or min-max scaling) helps optimisation algorithms converge and prevents attributes with large numeric ranges from dominating model calculations. Regularisation, by contrast, is implemented as an extra term in the loss function that imposes a penalty on model weights during training, with no direct relationship to feature magnitudes.
✓
Adding a penalty to the loss function to discourage overly complex models and reduce overfitting
Why this is correct
Adding a penalty to the loss function is exactly what regularisation accomplishes: the algorithm minimises both the original error and a penalty term such as L1 (sum of absolute weights) or L2 (sum of squared weights). This constraint pushes learned weights toward smaller values, thereby reducing model complexity and mitigating overfitting by improving generalisation to unseen data. The penalty strength is controlled by a hyperparameter (often λ), balancing fit versus simplicity.
✗
Applying government regulations to ensure AI models comply with data privacy laws
Why it's wrong here
Applying government regulations to ensure AI models comply with data privacy laws falls under responsible AI governance and legal compliance, not a numerical modelling technique. Regulations dictate organisational processes, documentation, and consent requirements, whereas regularisation is a purely mathematical adjustment to the objective function. Conf mixing regulatory compliance with regularisation reflects a categorical misunderstanding of terminology.
✗
Converting irregular training data shapes into a uniform format for the algorithm
Why it's wrong here
Converting irregular training data shapes into a uniform format, such as padding variable-length sequences or resizing images, is a data preparation step that ensures tensors have compatible dimensions for a model. It does not alter the loss function or impose constraints on weights. Regularisation is a model-level mechanism that modifies the training objective to prevent overfitting, whereas data reshaping is applied before training as part of the input pipeline.

Go deeper

AI-900 Practice Question: Describe fundamental principles of machine learning on Azure

Why each option matters

Option-by-option breakdown

Related to this question