What Does Evaluation of Explainable Artificial Intelligence Actually Tell Us? A Case for Compositional and Contextual Validation of XAI Building Blocks
This work addresses the problem of elusive evaluation in XAI for researchers and practitioners, offering a structured approach that is incremental in refining existing validation methods.
The paper tackles the challenge of evaluating explainable artificial intelligence (XAI) by proposing a fine-grained validation framework that separates technical building blocks, user-facing artefacts, and social protocols, aiming to systematically assess properties and downstream influence in view of anticipated use cases.
Despite significant progress, evaluation of explainable artificial intelligence remains elusive and challenging. In this paper we propose a fine-grained validation framework that is not overly reliant on any one facet of these sociotechnical systems, and that recognises their inherent modular structure: technical building blocks, user-facing explanatory artefacts and social communication protocols. While we concur that user studies are invaluable in assessing the quality and effectiveness of explanation presentation and delivery strategies from the explainees' perspective in a particular deployment context, the underlying explanation generation mechanisms require a separate, predominantly algorithmic validation strategy that accounts for the technical and human-centred desiderata of their (numerical) outputs. Such a comprehensive sociotechnical utility-based evaluation framework could allow to systematically reason about the properties and downstream influence of different building blocks from which explainable artificial intelligence systems are composed -- accounting for a diverse range of their engineering and social aspects -- in view of the anticipated use case.