Explainable AI (XAI) refers to a collection of methods aimed at making the inner workings and decision-making of AI models interpretable and understandable.
Unlike traditional black-box models, XAI enables the disclosure of influencing factors, decision rules, or probabilities that lead to a prediction.
Common techniques include LIME (Local Interpretable Model-agnostic Explanations), SHAP (SHapley Additive exPlanations), and saliency-based visualizations.
Explainable AI is a crucial research field to ensure trust, fairness, and accountability in AI systems.
Definition:
Explainable AI refers to methods and techniques aimed at making the decisions of AI models transparent, understandable, and interpretable for humans.
Source:
Adadi, A., & Berrada, M. (2018). "Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)." IEEE Access, 6, 52138-52160.