What is SHAP and why is it used in supervised learning?

Updated May 17, 2026

Short answer

SHAP explains model predictions by assigning contribution values to each feature.

Deep explanation

SHAP (SHapley Additive exPlanations) is based on cooperative game theory. It assigns each feature a contribution value representing its impact on prediction. SHAP ensures consistency and local interpretability, making it widely used in explainable AI.

Real-world example

Explaining why a bank rejected a loan application.

Common mistakes

  • Using SHAP without understanding baseline assumptions.

Follow-up questions

  • What is the difference between SHAP and feature importance?
  • Why is SHAP computationally expensive?

More Supervised Learning interview questions

View all →