Machine learning model interpretability using SHAP values: Application to a seismic facies classification task
David Lubo-Robles, Deepak Devegowda, Vikram Jayaram, Heather Bedle, Kurt J. Marfurt, Matthew J. Pranter
Abstract
Machine learning models have been widely used by geoscientists to accelerate their interpretation and highlight hidden patterns in their data. However, as the complexity of the model increases, the interpretation of the results can become quite challenging. The SHAP technique provides a measure of the importance of each of the input seismic attributes on the model’s output. We illustrate the value of the SHAP technique using a tree-based machine learning implementation trained to distinguish between Mass Transport Deposits (MTDs) and salt seismic facies in a Gulf of Mexico survey. Presentation Date: Monday, October 12, 2020 Session Start Time: 1:50 PM Presentation Time: 3:55 PM Location: 351F Presentation Type: Oral