Timezone: »

 
Poster
RKHS-SHAP: Shapley Values for Kernel Methods
Siu Lun Chau · Robert Hu · Javier González · Dino Sejdinovic

Tue Nov 29 02:00 PM -- 04:00 PM (PST) @ Hall J #433
in Poster Session 2 »

Feature attribution for kernel methods is often heuristic and not individualised for each prediction. To address this, we turn to the concept of Shapley values (SV), a coalition game theoretical framework that has previously been applied to different machine learning model interpretation tasks, such as linear models, tree ensembles and deep networks. By analysing SVs from a functional perspective, we propose RKHS-SHAP, an attribution method for kernel machines that can efficiently compute both Interventional and Observational Shapley values using kernel mean embeddings of distributions. We show theoretically that our method is robust with respect to local perturbations - a key yet often overlooked desideratum for consistent model interpretation. Further, we propose Shapley regulariser, applicable to a general empirical risk minimisation framework, allowing learning while controlling the level of specific feature's contributions to the model. We demonstrate that the Shapley regulariser enables learning which is robust to covariate shift of a given feature and fair learning which controls the SVs of sensitive features.

Keywords: Kernel Methods · Shapley values · Explainable AI
[ Paper]  [ Poster]  [ OpenReview] 

Author Information

Siu Lun Chau (University of Oxford)
Robert Hu (Amazon)
Javier González (Microsoft, Health Futures)
Dino Sejdinovic (University of Adelaide)

More from the Same Authors