Seminar on History and Philosophy of Science
Dabney Hall 110 (Treasure Room)
On Plausible Counterfactual Explanations in XAI
Jiji Zhang,
Professor of Philosophy and Associate Dean (Research), Faculty of Arts,
The Chinese University of Hong Kong,
Abstract: A popular approach in explainable artificial intelligence (XAI) is to generate so-called counterfactual explanations for outputs of predictive models. In this talk, I attempt a philosophical critique of key aspects of this approach and argue for a tension between two constraints on a good counterfactual explanation in this context, often known as 'plausibility' and 'causality'. I suggest a way to address this tension based on recent work on counterfactual generation.
For more information, please contact Fran Tise by phone at 626-395-3609 or by email at ftise@hss.caltech.edu.
Event Series
Seminar on History and Philosophy of Science