Expert-Informed, User-Centric Explanations for Machine Learning
Michael J. Pazzani, Severine Soltani, Robert Kaufman, Samson Qian, Albert Hsiao
Abstract
We argue that the dominant approach to explainable AI for explaining image classification, annotating images with heatmaps, provides little value for users unfamiliar with deep learning. We argue that explainable AI for images should produce output like experts produce when communicating with one another, with apprentices, and with novices. We provide an expanded set of goals of explainable AI systems and propose a Turing Test for explainable AI.
Topics & Concepts
Computer scienceArtificial intelligenceApprenticeshipTuring testSet (abstract data type)Value (mathematics)TuringTest (biology)Machine learningHuman–computer interactionLinguisticsBiologyPaleontologyProgramming languagePhilosophyExplainable Artificial Intelligence (XAI)Machine Learning in HealthcareArtificial Intelligence in Healthcare and Education