Abstracting General Syntax for XAI after Decomposing Explanation Sub-Components
Springer Artificial Intelligence (Springer AI), 2024
This work introduces the Qi-Framework, a mathematically grounded syntax for describing explainability requirements across machine learning systems. By decomposing common eXplainable AI (XAI) techniques into modular sub-components, the framework standardizes how practitioners reason about their explanation needs, compare methods, and identify gaps in available tooling. The framework supports ranking XAI approaches by utility for a target use case and encourages collaborative development of interpretable AI techniques.
@article{wormald2024abstracting,
title={Abstracting General Syntax for XAI after Decomposing Explanation Sub-Components},
author={Wormald, Stephen and Maldaner, Matheus Kunzler and O'Connor, Kristian and Dizon-Paradis, Olivia P. and Woodard, Damon L.},
journal={Artificial Intelligence},
publisher={Springer},
year={2024}
}