COMPEL Glossary / explainable-ai-xai
Explainable AI (XAI)
Explainable AI (XAI) is a field of research and practice focused on developing techniques, tools, and methodologies that make AI decision-making processes understandable to humans.
What this means in practice
XAI addresses the 'black box' problem where complex models (particularly deep learning) produce outputs through mathematical transformations that resist straightforward interpretation. XAI techniques include LIME (Local Interpretable Model-agnostic Explanations, which approximates model behavior locally), SHAP (SHapley Additive exPlanations, which attributes importance to each input feature), attention visualization (showing what parts of input the model focused on), and counterfactual explanations (what would need to change for a different outcome). XAI is increasingly required by regulations like the EU AI Act for high-risk AI systems and is a key component of the transparency requirements in COMPEL Domain 15.
Related Terms
Other glossary terms mentioned in this entry's definition and context.