Explaining how to explain our underwriting model at the local data science meetup

I’m going to start posting to this page with professional news and writing snippets. Here is an outdated update from October: I gave a meetup talk on how we are using additive feature attribution to make our algorithmic underwriting process more transparent at MassMutual.

I went over the current state of research on local explanation methods and how we are starting to use SHAP values as a jumping-off point for learning about & debugging our model, explaining results to stakeholders across the company, and providing information to applicants.