Explain bias-variance, calibration, and model drift

You are interviewing for an applied ML role. Answer the following ML fundamentals questions in a business-facing way (i.e., start from a customer/business need, then map it to the ML concept and actions).

1) Bias–variance trade-off

• Define bias and variance and explain the trade-off.
• Give practical ways to reduce bias .
• Give practical ways to reduce variance .

2) Confidence (probability) calibration

• Define confidence calibration (what does it mean for a model to be calibrated?).
• How would you measure calibration?
• What are common methods to improve calibration?

3) Model drift

• Define model drift (and distinguish common types of drift).
• How would you detect drift in production?
• What are typical mitigation actions once drift is detected?

Assume a typical product scenario (e.g., ranking/recommendation, fraud detection, churn prediction, ads CTR) where the customer wants stable, reliable decisions and probabilities over time.