Machine Learning / 18
Explainability
Study global and local explanations, feature importance, and the limits of interpreting model behavior.
An explanation can describe model behavior. It does not automatically reveal causation.
This is a knowledge index. Published titles open their articles; the remaining entries outline planned coverage.
Subtopics
- 18.01
Explainability vs Interpretability
Planned guide: Explainability vs Interpretability. Focus on intuition, assumptions, and a worked example.
- 18.02
Global Explanations
Planned guide: Global Explanations. Focus on intuition, assumptions, and a worked example.
- 18.03
Local Explanations
Planned guide: Local Explanations. Focus on intuition, assumptions, and a worked example.
- 18.04
Linear Model Coefficients
Planned guide: Linear Model Coefficients. Focus on intuition, assumptions, and a worked example.
- 18.05
Feature Importance
Planned guide: Feature Importance. Focus on intuition, assumptions, and a worked example.
- 18.06
Permutation Importance
Planned guide: Permutation Importance. Focus on intuition, assumptions, and a worked example.
- 18.07
Partial Dependence
Planned guide: Partial Dependence. Focus on intuition, assumptions, and a worked example.
- 18.08
ICE Plots
Planned guide: ICE Plots. Focus on intuition, assumptions, and a worked example.
- 18.09
SHAP
Explore SHAP explanations and the limits of interpreting feature contributions.
- 18.10
LIME
Planned guide: LIME. Focus on intuition, assumptions, and a worked example.
- 18.11
Counterfactual Explanations
Planned guide: Counterfactual Explanations. Focus on intuition, assumptions, and a worked example.
- 18.12
Explanation Limitations
Planned guide: Explanation Limitations. Focus on intuition, assumptions, and a worked example.