Skip to yearly menu bar Skip to main content


Regularizing Black-box Models for Improved Interpretability

Gregory Plumb · Maruan Al-Shedivat · Ángel Alexander Cabrera · Adam Perer · Eric Xing · Ameet Talwalkar

Poster Session 3 #1078

Keywords: [ Deep Learning ] [ Algorithms -> Density Estimation; Algorithms -> Uncertainty Estimation; Algorithms -> Unsupervised Learning; Deep Learning ] [ G ]


Most of the work on interpretable machine learning has focused on designing either inherently interpretable models, which typically trade-off accuracy for interpretability, or post-hoc explanation systems, whose explanation quality can be unpredictable. Our method, ExpO, is a hybridization of these approaches that regularizes a model for explanation quality at training time. Importantly, these regularizers are differentiable, model agnostic, and require no domain knowledge to define. We demonstrate that post-hoc explanations for ExpO-regularized models have better explanation quality, as measured by the common fidelity and stability metrics. We verify that improving these metrics leads to significantly more useful explanations with a user study on a realistic task.

Chat is not available.