X-SHAP: towards multiplicative explainability of Machine Learning
This provides a tool for explainable AI in domains where multiplicative interactions are critical, such as insurance and biology, though it is an incremental extension of existing SHAP methods.
The paper introduces X-SHAP, a model-agnostic method that extends additive SHAP to assess multiplicative variable contributions in predictions, addressing multiplicative interactions common in fields like insurance and biology. It tests the method on various datasets and compares it to traditional techniques for capturing multiplicative feature importance.
This paper introduces X-SHAP, a model-agnostic method that assesses multiplicative contributions of variables for both local and global predictions. This method theoretically and operationally extends the so-called additive SHAP approach. It proves useful underlying multiplicative interactions of factors, typically arising in sectors where Generalized Linear Models are traditionally used, such as in insurance or biology. We test the method on various datasets and propose a set of techniques based on individual X-SHAP contributions to build aggregated multiplicative contributions and to capture multiplicative feature importance, that we compare to traditional techniques.