Explaining Deep Learning Models for Tabular Data Using Layer-Wise Relevance Propagation

Trust and credibility in machine learning models are bolstered by the ability of a model to explain its decisions. While explainability deep is well-known challenge, further challenge clarity explanation itself for relevant stakeholders model. Layer-wise Relevance Propagation (LRP), an established technique developed computer vision, provides intuitive human-readable heat maps input images. We present novel application LRP with tabular datasets containing mixed data (categorical numerical) using neural network (1D-CNN), Credit Card Fraud detection Telecom Customer Churn prediction use cases. show how more effective than traditional concepts Local Interpretable Model-agnostic Explanations (LIME) Shapley Additive (SHAP) explainability. This effectiveness both local sample level holistic over whole testing set. also discuss significant computational time advantage (1–2 s) LIME (22 SHAP (108 on same laptop, thus potential real scenarios. In addition, our validation has highlighted features enhancing performance, opening up new area research XAI as approach feature subset selection.