Speaker: Ian Covert (University of Washington)

Location: Click here to join the meeting (Microsoft Teams Meeting)

Title: SAGE, a principled approach to calculating global feature importance using Shapley values

Ian Covert's talk will be based on his recent paper: https://arxiv.org/abs/2004.00668

Summary:
Understanding the inner workings of complex machine learning models is a long-standing problem and most recent research has focused on local interpretability. To assess the role of individual input features in a global sense, we explore the perspective of defining feature importance through the predictive power associated with each feature. We introduce two notions of predictive power (model-based and universal) and formalize this approach with a framework of additive importance measures, which unifies numerous methods in the literature. We then propose SAGE, a model-agnostic method that quantifies predictive power while accounting for feature interactions. Our experiments show that SAGE can be calculated efficiently and that it assigns more accurate importance values than other methods.

This seminar is organised together with XAI Norway.