Provably Robust Model-Centric Explanations for Critical Decision-Making
Machine Learning
2021-10-28 v1 Artificial Intelligence
Robotics
Abstract
We recommend using a model-centric, Boolean Satisfiability (SAT) formalism to obtain useful explanations of trained model behavior, different and complementary to what can be gleaned from LIME and SHAP, popular data-centric explanation tools in Artificial Intelligence (AI). We compare and contrast these methods, and show that data-centric methods may yield brittle explanations of limited practical utility. The model-centric framework, however, can offer actionable insights into risks of using AI models in practice. For critical applications of AI, split-second decision making is best informed by robust explanations that are invariant to properties of data, the capability offered by model-centric frameworks.
Cite
@article{arxiv.2110.13937,
title = {Provably Robust Model-Centric Explanations for Critical Decision-Making},
author = {Cecilia G. Morales and Nicholas Gisolfi and Robert Edman and James K. Miller and Artur Dubrawski},
journal= {arXiv preprint arXiv:2110.13937},
year = {2021}
}
Comments
8 pages, 9 figures