A Unified Approach to Interpreting Model Predictions, Scott M. Lundberg and Su-In Lee, 2017Advances in Neural Information Processing Systems (NeurIPS 30), Vol. 30 (Curran Associates, Inc.)DOI: 10.5555/3295222.3295230 - Presents the SHAP framework, which unifies various interpretability methods and provides a theoretically sound approach for feature attribution in models, including regression.
SHAP Documentation, Scott Lundberg and SHAP Contributors, 2024 - The official documentation for the SHAP Python library, providing usage examples, API reference, and details relevant to applying SHAP to regression models.
LIME: Local Interpretable Model-agnostic Explanations Documentation, Marco Ribeiro, Sameer Singh, Carlos Guestrin, and LIME Contributors, 2024 - The official documentation for the LIME Python library, offering practical guidance and code examples for explaining individual regression predictions.