Explainability and Analysis of Variance (Zijun Gao, USC)

Abstract: Existing tools for explaining complex models and systems are associational rather than causal and do not provide mechanistic understanding. We propose a new notion called counterfactual explainability for causal attribution that is motivated by the concept of genetic heritability in twin studies. Counterfactual explainability extends methods for global sensitivity analysis (including the functional analysis of variance and Sobol’s indices), which assumes independent explanatory variables, to dependent explanations by using a directed acyclic graphs to describe their causal relationship. Therefore, this explanability measure directly incorporates causal mechanisms by construction. Under a comonotonicity assumption, we discuss methods for estimating counterfactual explainability and apply them to a real dataset dataset to explain income inequality by gender, race, and educational attainment.

Bio: Zijun Gao is a tenure-track assistant professor in the Department of Data Sciences and Operations at USC Marshall Business School. She received her Ph.D. in Statistics from Stanford University in 2022 supervised by Professor Trevor Hastie. She served as a research associate in the Statistical Lab at the University of Cambridge from 2022 to 2023 hosted by Professor Qingyuan Zhao. Her research focuses on the estimation and inference problems in causal inference with heterogeneity, with side interests in distribution learning, selective inference, and model evaluation. She also works on real-world data motivated topics, with a specific emphasis on the applications in adaptive clinical trial and personalized medication.