Interpretability

Caesar AI Atlas Definition

Interpretability is the extent to which a human can understand how an AI system or model produces its outputs. It may come from inherently understandable model structures or from methods that explain features, rules, reasoning paths, or output drivers.

Other Definitions

Interpretability Source

The ability to explain or to present an ML model's reasoning in understandable terms to a human. Most linear regression models, for example, are highly interpretable. (You merely need to look at the trained weights for each feature.) Decision forests are also highly interpretable. Some models, however, require sophisticated visualization to become interpretable. You can use the Learning Interpretability Tool (LIT) to interpret ML models.

Interpretability Source

Some machine learning models, particularly those trained with deep learning, are so complex that it may be difficult or impossible to know how the model produced the output (PB 57, PN 633). Interpretability often describes the ability to present or explain a machine learning system's decision-making process in terms that can be understood by humans (PN 633). Interpretability is sometimes referred to as transparency or explainability.

Also Referenced In

basebox — AI Glossary for Executives

Concept Comparisons

Common ConfusionAi Governance

Explainable AI vs Interpretability

A side-by-side comparison of Explainable AI and Interpretability. Understand how explanation-oriented systems and methods relate to the human ability to understand model behavior.