Position: In Defence of Post-hoc Explainability
Keywords: Computational Interpretabilism, Explainable AI, Post-hoc Explanations, AI Ethics, Machine Learning Interpretability, Epistemology of AI, AI Transparency, Black Box Models
TL;DR: This paper challenges prevailing critiques of post-hoc explanations in AI systems.
Abstract: The widespread adoption of machine learning in scientific research has created a fundamental tension between model opacity and scientific understanding. Whilst some advocate for intrinsically interpretable models, we introduce Computational Interpretabilism (CI) as a philosophical framework for post-hoc interpretability in scientific AI. Drawing parallels with human expertise, where post-hoc rationalisation coexists with reliable performance, CI establishes that scientific knowledge emerges through structured model interpretation when properly bounded by empirical validation. Through mediated understanding and bounded factivity, we demonstrate how post-hoc methods achieve epistemic justification without requiring complete mechanical transparency, resolving tensions between model complexity and scientific comprehension.
Track: Position paper track
Submitted Paper: No
Published Paper: No
Submission Number: 72
Loading