Feature Scaling — Google Cloud ML Engineer Practice Questions

Feature scaling is the broader category of techniques that transform numeric features to a common scale so that model training is numerically stable and faster. The Google Cloud ML Engineer exam covers two primary approaches: min-max normalization, which maps values to a fixed range, and standardization, which centers values at zero with unit variance. Candidates must recognize which algorithms require scaling, such as logistic regression, support vector machines, and neural networks, versus tree-based algorithms that are scale-invariant, and must understand how to apply consistent transformations at training and serving time to avoid training-serving skew.

Free questions on feature scaling

More feature scaling questions in the full bank

• What is the primary difference between normalization and standardization? Unlock answer & explanation →