Poster
Explaining Preferences with Shapley Values
Robert Hu · Siu Lun Chau · Jaime Ferrando Huertas · Dino Sejdinovic
Hall J (level 1) #905
Keywords: [ interpretability ] [ Shapley values ] [ RKHS ] [ kernel ] [ preference learning ]
While preference modelling is becoming one of the pillars of machine learning, the problem of preference explanation remains challenging and underexplored. In this paper, we propose \textsc{Pref-SHAP}, a Shapley value-based model explanation framework for pairwise comparison data. We derive the appropriate value functions for preference models and further extend the framework to model and explain \emph{context specific} information, such as the surface type in a tennis game. To demonstrate the utility of \textsc{Pref-SHAP}, we apply our method to a variety of synthetic and real-world datasets and show that richer and more insightful explanations can be obtained over the baseline.