Machine-learning coding exercise: build a regression model on numerical features that extrapolates well beyond the training range

Design and implement a regression solution (code expected) that not only fits the training data but also maintains low error when test points fall outside the feature ranges seen in training. Explain feature engineering, model choice, regularization, and how you will evaluate extrapolation performance.

Consider linear or monotonic models, polynomial basis with regularization, data standardization, and a hold-out test split drawn from an expanded feature range.

Upstart Data Scientist interview question: Design a Regression Model for Robust Extrapolation Performance. {"blocks": [{"key": "a4f84778", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {...