SHAP (SHapley Additive exPlanations)

SHAP is a game theory-based approach to explain the output of any machine learning model. It assigns each input feature an importance value showing how much it contributed to a specific prediction.

SHAP explains which inputs drove a model's output and by how much, turning an opaque score into an attributable one. It is defined here because you will meet it in vendor documentation and research literature.

Further reading: Wikipedia