Additive-feature-attribution methods: a review on explainable artificial intelligence for fluid dynamics and heat transfer

TL;DR

This paper reviews additive-feature-attribution methods, especially SHAP, highlighting their role in interpreting deep learning models in fluid dynamics and heat transfer, emphasizing their importance for physics-compliant explainability.

Contribution

It provides a comprehensive overview of additive-feature-attribution methods and their applications in fluid mechanics, emphasizing their significance for interpretable AI models.

Findings

Four main implementations: kernel SHAP, tree SHAP, gradient SHAP, deep SHAP.

Applications span turbulence modeling, fundamentals, and applied fluid dynamics.

Explainability techniques are vital for physics-aware deep learning in fluid mechanics.

Abstract

The use of data-driven methods in fluid mechanics has surged dramatically in recent years due to their capacity to adapt to the complex and multi-scale nature of turbulent flows, as well as to detect patterns in large-scale simulations or experimental tests. In order to interpret the relationships generated in the models during the training process, numerical attributions need to be assigned to the input features. One important example are the additive-feature-attribution methods. These explainability methods link the input features with the model prediction, providing an interpretation based on a linear formulation of the models. The SHapley Additive exPlanations (SHAP values) are formulated as the only possible interpretation that offers a unique solution for understanding the model. In this manuscript, the additive-feature-attribution methods are presented, showing four common implementations in the literature: kernel SHAP, tree SHAP, gradient SHAP, and deep SHAP. Then, the main applications of the additive-feature-attribution methods are introduced, dividing them into three main groups: turbulence modeling, fluid-mechanics fundamentals, and applied problems in fluid dynamics and heat transfer. This review shows thatexplainability techniques, and in particular additive-feature-attribution methods, are crucial for implementing interpretable and physics-compliant deep-learning models in the fluid-mechanics field.