How do you justify shipping logistic regression over a boosted ensemble that scores better?
mediumAnswer
- Frame it as total cost, not offline metric.
- If the gap is a fraction of a point of AUC and the decision threshold sits in a region where both models rank the same customers, the business outcome is identical.
- Logistic regression gives you coefficients a regulator can read, near-zero inference latency, trivial monitoring, and calibrated probabilities out of the box.
- It also fails predictably under drift instead of in surprising ways.
- Quantify the gap in the metric the business actually uses, and if it is negligible, the simpler model is the better engineering decision.
Check yourself — multiple choice
- Always ship the highest-AUC model
- Translate the gap into business impact; if negligible, interpretability, latency, calibration and predictable failure win
- Simpler models are never justified
- Only regulators matter
Compare total cost and business impact rather than the raw offline metric.
#interview-scenarios#interpretability
Practise Supervised Learning
215 interview questions in this topic.
Related questions
- What are the limits of SHAP values when explaining a model to a stakeholder?
- How do you interpret a coefficient in a multiple linear regression?
- How do you interpret a logistic regression coefficient as an odds ratio?
- How is feature importance computed from a decision tree or random forest?
- What are monotonic constraints in XGBoost / LightGBM and when do you use them?
- You must pick ONE metric for your model. How do you decide?