CQD-SHAP: Explainable Complex Query Answering via Shapley Values
This episode examines the paper 'CQD-SHAP: Explainable Complex Query Answering via Shapley Values' by P. Abbasi and S. Heindorf. The hosts discuss using Shapley values to provide deep attribution for AI answers in knowledge graphs, focusing on reducing computational overhead and translating mathematical insights into intuitive, natural language for users.