A unified framework for evaluating the robustness of machine-learning interpretability for prospect risking
This addresses the need for more reliable interpretability in geophysics for hydrocarbon exploration, though it is incremental as it builds on existing XAI methods.
The paper tackles the problem of inconsistent explanations from XAI methods like LIME and SHAP in hydrocarbon prospect risking by proposing a unified framework that uses counterfactuals and causal concepts of necessity and sufficiency to evaluate robustness, providing insights into model capabilities and optimal XAI-model pairings for the dataset.
In geophysics, hydrocarbon prospect risking involves assessing the risks associated with hydrocarbon exploration by integrating data from various sources. Machine learning-based classifiers trained on tabular data have been recently used to make faster decisions on these prospects. The lack of transparency in the decision-making processes of such models has led to the emergence of explainable AI (XAI). LIME and SHAP are two such examples of these XAI methods which try to generate explanations of a particular decision by ranking the input features in terms of importance. However, explanations of the same scenario generated by these two different explanation strategies have shown to disagree or be different, particularly for complex data. This is because the definitions of "importance" and "relevance" differ for different explanation strategies. Thus, grounding these ranked features using theoretically backed causal ideas of necessity and sufficiency can prove to be a more reliable and robust way to improve the trustworthiness of the concerned explanation strategies.We propose a unified framework to generate counterfactuals as well as quantify necessity and sufficiency and use these to perform a robustness evaluation of the explanations provided by LIME and SHAP on high dimensional structured prospect risking data. This robustness test gives us deeper insights into the models capabilities to handle erronous data and which XAI module works best in pair with which model for our dataset for hydorcarbon indication.